Assessing skin lesion evolution from multispectral image sequences

Sylvain Prigent 1 Xavier Descombes 1 Didier Zugaj 2 Laurent Petit 2 Anne-Sophie Dugaret 2 Philippe Martel 2 Josiane Zerubia 3
1 MORPHEME - Morphologie et Images
CRISAM - Inria Sophia Antipolis - Méditerranée , IBV - Institut de Biologie Valrose : U1091, Laboratoire I3S - SIS - Signal, Images et Systèmes
Abstract : During the evaluation of skin disease treatments, dermatologists have to clinically measure the evolution of the pathology severity of each patient during treatment periods. Such a process is sensitive to intra- and inter- dermatologist diagnosis. To make this severity measurement more objective we quantify the pathology severity using a new image processing based method. We focus on a hyperpigmentation disorder called melasma. During a treatment period, multispectral images are taken on patients receiving the same treatment. After co-registration and segmentation steps, we propose an algorithm to measure the intensity, the size and the homogeneity evolution of the pathological areas. Obtained results are compared with a dermatologist diagnosis using statistical tests on two clinical studies containing respectively 384 images from 16 patients and 352 images from 22 patients. This research report is an update of the report 8136. It describes methods and experiments in more details and provides more references.
Document type :
Reports
Complete list of metadatas

Cited literature [63 references]  Display  Hide  Download

https://hal.inria.fr/hal-01164502
Contributor : Sylvain Prigent <>
Submitted on : Monday, June 22, 2015 - 11:54:36 AM
Last modification on : Monday, November 5, 2018 - 3:52:02 PM
Long-term archiving on : Tuesday, September 15, 2015 - 8:26:57 PM

File

RR-8745.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01164502, version 1

Collections

Citation

Sylvain Prigent, Xavier Descombes, Didier Zugaj, Laurent Petit, Anne-Sophie Dugaret, et al.. Assessing skin lesion evolution from multispectral image sequences. [Research Report] RR-8745, Inria Sophia Antipolis; INRIA. 2015. ⟨hal-01164502⟩

Share

Metrics

Record views

406

Files downloads

372