Rapid three-dimensional measuring system for facial surface structure.

Second Department of Oral and Maxillofacial Surgery, Faculty of Dentistry, Osaka University, Suita, Japan.
Plastic &amp Reconstructive Surgery (Impact Factor: 3.54). 12/1998; 102(6):2108-13.
Source: PubMed

ABSTRACT A noncontact three-dimensional measuring system (liquid crystal range finder system) is described. Three-dimensional facial surface data (more than 30,000 points) could be obtained in 1 second, and the resolution was approximately 0.4 mm. The reliability and repeatability of the results were validated with a calibrating apparatus and a highly accurate contact-type three-dimensional digitizer. Consequently, the average of the measurement errors on a facial plaster model was 0.3 mm. Repeatability in measuring human faces was approximately 0.3 mm. Therefore, the total error in measuring human faces was approximately 0.5 mm. Because of the shortness of measuring time, this system was capable of scanning faces of infants without the need for sedation. The output of the liquid crystal range finder was demonstrated on an infant with cleft lip. The surface points improved by cheiloplasty, and the residual deformities were observed clearly. This system was thought to be the most suitable apparatus for measuring faces of infants (especially infants with cleft lip) and enabled us to analyze facial surface structure both qualitatively and quantitatively.

  • [Show abstract] [Hide abstract]
    ABSTRACT: 3D digital surface photogrammetry is an objective means of documenting the quantitative evaluation of facial morphology. However, there are no standardized superimposition and measurement systems for surveying soft tissue changes. The aim of this study was to present a superimposition and measurement model for three-dimensional analysis of therapy-induced sagittal changes in facial soft tissue and to ascertain its applicability based on the reproducibility of 3D landmark positions. Twenty-nine children were examined (eight with cleft lip and palate, six with cleft palate, eight with Class III malocclusion and seven healthy controls, between 4.1 and 6.4 years). The mean time between examinations was 8.2 months for the patients and 8 months for the control group. Data was acquired with the DSP 400((c))imaging system. A mathematical model with seven superimposition points was developed. Two 3D images, one at the beginning and the other at the end of the examination, were generated. Both images were superimposed ten times. Ten landmarks for evaluating the soft tissue changes were geometrically defined on the superimposition image, put in place ten times, and measured. The landmarks' reproducibility was calculated via statistical intraoperator analysis. Measurement error was identified using the root mean square error (RMSE). The superimposition points were easy to locate and the landmarks well definable. All midface landmarks proved to be highly reproducible with an RMSE under 0.50 mm. The lower face landmarks demonstrated good reproducibility with an RMSE under 1 mm. The midface landmarks' precision fell below the range of accuracy, while the lower face landmarks' precision fell within the optoelectronic scanner device's range of accuracy (0.50-1 mm). As an accurate, non-invasive, millisecond-fast, non-ionizing and ad infinitum repeatable procedure, 3D digital surface photogrammetry is very well suited for clinical and scientific application in orthodontics. We developed a reliable superimposition and measurement model with 3D digital surface photogrammetry. This new capturing and measurement system provides a simple means of determining 3D changes in facial soft tissue. Our landmarks proved to be highly reproducible for the midface while revealing good reproducibility for the lower face.
    Fortschritte der Kieferorthopädie 05/2010; 71(3):221-34. · 0.89 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Prevalence and incidence rates of lipodystrophy vary widely and frequently rely upon self- and/or clinician reports. Currently no validated assessment tool for facial lipoatrophy is available. To illustrate that assessment of the severity of facial lipoatrophy by patients and clinicians is subjective. To evaluate the reproducibility of facial three-dimensional surface laser scans. Twenty-three HIV-positive men were recruited from an inner London HIV outpatient clinic in September 2001. CD4 count, viral load, antiretroviral history and body mass index were recorded. Patients and clinicians independently assessed the severity of facial lipoatrophy on a four-point scale and the level agreement was measured. Seventeen of the 23 patients (73.9%) underwent two scans 1 week apart, which were then superimposed. The volume difference (mm3) and mean difference (mm) between the scans for five regions of the face were measured and compared with the self-reported grade of facial lipoatrophy. For each pair of clinicians (P=0.03, 0.005 and 0.0002, respectively), and for one patient-clinician pair (P=0.004), there was a significant systematic difference between the two sets of gradings of facial lipoatrophy. The level of disagreement was generally higher for patients reporting facial lipoatrophy (n=17) compared to those not reporting it (n=6). The mean volume difference and mean difference between any region were within 200 mm3 and 0.25 mm, respectively. Reproducibility was unaffected by the self-reported grade of facial lipoatrophy. Assessment of the severity of facial lipoatrophy by patients and clinicians is subjective. Three-dimensional facial laser scans are reproducible and may provide an objective tool for monitoring changes in facial lipoatrophy.
    HIV Medicine 11/2003; 4(4):325-31. · 3.16 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Advances in technology have yielded systems that are capable of non-invasive, high-definition three-dimensional recordings of human body surfaces. Originally designed for use in an industrial setting, surface scanners have not been sufficiently evaluated for medical application so far. In this study six scanners based on different scanning principles (ringe projection scanners, laser scanners and photogrammetry scanner) were employed to measure five sheep skulls of different sizes. Distance measurements were done chephalometric or on the three dimensional data sets and compared concerning the accuracy. Conclusion of this work is with all scanning principles a skull can be recorded in seconds, archived and distances measured with the accuracy of a 1/10 millimeter. Best results will be achieved when the different types of scanners are applied to specific medical problems according to their special construction designs and characteristics.