Premium
Critical study of hair growth analysis with computer‐assisted methods
Author(s) -
Van Neste D,
Trüeb RM
Publication year - 2006
Publication title -
journal of the european academy of dermatology and venereology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.655
H-Index - 107
eISSN - 1468-3083
pISSN - 0926-9959
DOI - 10.1111/j.1468-3083.2006.01568.x
Subject(s) - scalp , medicine , hair growth , biomedical engineering , image processing , computer vision , software , artificial intelligence , pathology , computer science , dermatology , image (mathematics) , physiology , programming language
Background Computer‐assisted image analysis has been proposed for human hair growth studies. Methods The performances of Trichoscan, a commercially available automated system combining epiluminiscence microscopy with digital image analysis, developed for office‐based hair growth measurements, have been evaluated comparatively on the same skin sites using standardized photographic equipment and calibrated processing for contrast‐enhanced phototrichogram (CE‐PTG) analysis. This reference method has been validated with scalp biopsies and histological examination of serial sectioning. Results Besides edge effects, hair fibres escaped the Trichoscan analysis for various reasons including, but not limited to, thickness, pigmentation, closeness and crossing. Conclusion Most of these problems have been identified in the late 1980s and remain largely unsolved by the processing software that was evaluated in 2004. Therefore claims promoting the Trichoscan method for accurate hair measurements in clinical trials on scalp and body hair are not supported by the present investigation. The speed at which the analysis is performed is outweighed by the errors in signal detection. Therefore we suggest that improvements must be clearly documented before Trichoscan is established for quantified diagnostic purposes and detailed hair cycle monitoring during hair trials.