- Email: [email protected]
TrichoScan: A Novel Tool For The Analysis of Hair Growth In Vivo
ORIGINAL ARTICLE
TrichoScan: A Novel Tool For The Analysis of Hair Growth In Vivo Rolf Ho¡mann Department of Dermatology, Philipp University, Marburg, Germany
Hair loss or hair thinning is a common complaint in clinical dermatology, and patients seeking advice for hair loss are not necessarily bald. Because the e¡ects of treatment attempts are hard to measure, there is need for a sensitive tool to monitor hair loss and treatment responses. Such a method must be able to analyze the biologic parameters of hair growth, which are: (i) hair density (n per cm2); (ii) hair diameter (lm); (iii) hair growth rate (mm per day); and (iv) anagen/telogen ratio. Herein we present the TrichoScan as a method that combines epiluminescence microscopy with automatic digital image analysis for the measurement of human, and potentially animal hair, in situ. The TrichoScan is able to analyze all four parameters of hair growth with a so-called intraclass correlation of approximately 91% within the same TrichoScan operator and an intraclass
correlation of approximately 97% for di¡erent TrichoScan operators. The application of the technique is demonstrated by comparing the hair parameters in individuals without apparent hair loss, men with untreated androgenetic alopecia, and men after treatment with ¢nasteride (1 mg per day). We were able to detect a signi¢cant increase in hair counts and cumulative hair thickness 3 and 6 mo after treatment. Advantages of the TrichoScan are that it can be used for clinical studies to compare placebo versus treatment, to compare di¡erent capacities of hair growth promoting substances, to study androgenetic alopecia and other forms of di¡use hair loss, and to study the e¡ects of drugs and laser treatment on hypertrichosis and hirsutism. Key words: alopecia/computer analysis/hair/measurement/treatment. JID Symposium Proceedings 8:109 ^115, 2003
H
(Rushton et al, 1993; Van Neste, 1999). Therefore, an operatorfriendly and patient-friendly, inexpensive, validated, and reliable method is a rational need. A useful method must be able to analyze the biologic parameters of hair growth, which are: (i) hair density (n per cm2); (ii) hair diameter (mm); (iii) hair growth rate (mm per day); and (iv) anagen/telogen ratio. This paper describes the TrichoScan as such a method, which combines standard epiluminescence microscopy (ELM) with automatic digital image analysis for the measurement of human, and potentially animal hair, in situ. Application of the technique is demonstrated by comparing the four hair parameters in individuals without apparent hair loss, men with untreated AGA, and men after treatment with ¢nasteride (1 mg per day).
air loss or hair thinning is a common complaint in clinical dermatology, and patients seeking advice for hair loss are not necessarily bald. In established cases of androgenetic alopecia (AGA) characteristic patterns are easily discernible; however, especially in women, the clinician is often challenged by patients with initial stages of AGA where hair loss is reported but alopecia is not recognizable or the e¡ects of treatment attempts are hard to measure. Consequently, there is a need for a sensitive tool to monitor hair loss and treatment responses. Numerous methods have been reported (Barth and Rushton, 1995) to assess rates of hair growth. These techniques can be classi¢ed as either (i) invasive, such as biopsies (Headington, 1984; Whiting et al, 1999); (ii) semi-invasive, such as the trichogram (Maguire and Kligman, 1964; Blume-Peytavi and Orfanos, 1995) and unit area trichogram (Rushton et al, 1983); or (iii) noninvasive, such as global hair counts (Can¢eld, 1996) and the phototrichogram (Saitoh et al, 1970; Bouhanna, 1985; Guarrera and Ciulla, 1986; Friedel et al, 1989; Van Neste et al, 1989, 1994). Quantitative methods to analyze human hair growth and hair loss are necessary to determine the e⁄cacy of hair promoting drugs. In reviewing capabilities of the di¡erent methods, a common theme emerges, that most techniques are of little use to the clinician because they are time consuming, costly, and/or di⁄cult to perform
MATERIALS AND METHODS Volunteers and patients A total of 56 persons (25 women, age range 25^48 y, mean 34 y; 31 men, age range 26^39 y, mean 32 y) were studied. Ten of 56 volunteers (¢ve women, age range 25^48 y, mean 34 y; ¢ve men, age range 26^39 y, mean 32 y) were recruited for the initial experiments to analyze the reproducibility of the method. The measured parameters were hair thickness and hair numbers at the occiput. Seventeen male patients (age range 25^48 y, mean 34 y), who had recognized progressive thinning of hair and hair loss for more than 4 y were included to analyze the progression of AGA, with and without treatment. All patients presented the clinical ¢nding of mild to moderate AGA with various degrees of involvement that were classi¢ed according to the Hamilton scale (II^V) (Hamilton, 1951). Patients with other forms of alopecia were excluded. Twelve of the 17 male patients with AGA were treated with Propecias (Merch & Co., Inc. Whitehouse Station, Rahway, NJ) (1 mg ¢nasteride per day) for 6 mo. All patients actively treated had
Manuscript accepted for publication February 1, 2003 Reprint requests to: Rolf Ho¡mann MD, Department of Dermatology, Philipp University, Deutschhausstrasse 9, 35033 Marburg, Germany. Email: [email protected]. Abbreviations: AGA, androgenetic alopecia; ELM, epiluminescence microscopy.
0022-202X/03/$15.00 . Copyright r 2003 by The Society for Investigative Dermatology, Inc. 109
110
HOFFMANN
had no treatment whatsoever for hair loss for at least 1 mo before initiating the study. Eleven of 56 healthy male volunteers (age range 28^55 y, mean 36 y) who had experienced no episodes of hair thinning or hair loss, recent illnesses, or general health disturbances, were recruited as a control group. Clinical examination revealed no evidence of any hair disorder with either the female or the male volunteers. The measured parameters were hair thickness and hair numbers at the vertex. For the analysis of daily hair growth and the anagen/telogen ratio, an additional 18 volunteer patients with AGA were recruited.
Clipping of hairs In individuals a¡ected by AGA, a transitional area of hair loss between normal hair and the balding area was de¢ned and an area of 1.8 cm2 was clipped (Hairliner, Wella Germany Darmstadt, Germany) (Fig 1A^D). In volunteers without AGA (controls) the vertex was chosen for clipping. All clipped areas were landmarked with a central, single black tattoo. The tattoo was visible throughout the study. In the 18 volunteers who were recruited for the analysis of the anagen/telogen ratios, the scalp was clipped at two locations (vertex and occiput) and was analyzed by two investigators with the TrichoScan software.
JID SYMPOSIUM PRECEEDINGS
ELM of clipped hairs Gray or fair hairs have only limited contrast in comparison with the scalp (Van Neste, 2001). Therefore, the clipped hairs within the target area were dyed for 12 min (Fig 1E^H) with a commercially available solution (RefectoCils, Gschwentner, Vienna, Austria), which is normally used to color eye brows and lashes. The approach of dying hairs for hair growth studies has been reported to give the same results as uncolored hairs (Pecoraro et al, 1971). For the analysis of hair number and thickness, hairs were colored immediately after shaving and for the analysis of the hair growth rate and anagen/telogen ratio, hairs were colored 3 d after shaving. Thereafter, the colored area was cleaned (Fig 1I) with an alcoholic solution (Kodans Spray, Schˇlke & Mayr, Vienna, Austria) and digital images were obtained at 20-fold (analyzed area: 0.62 cm2) and 40-fold (analyzed area: 0.225 cm2) magni¢cation by means of a digital ELM system (Foto¢nder DERMA, Teachscreen Software, Bad Birnbach, Germany) while the area was still wet (Fig 1J). This digital camera was equipped with a rigid contact lens that ensured images were always taken at the same distance from the scalp. Because the camera must be pressed on to the scalp, the hairs were always £attened. Images were taken at day 0, immediately after clipping, 2 and 3 d after clipping, and 3 and 6 mo after the initial visit, respectively. Two di¡erent investigators each took three images from the same patient at every visit. Software for digital image and statistical analysis Software was developed to analyze (TrichoScan) hair density (n per cm2), hair diameter (mm), hair growth rate (mm per day), and anagen/telogen ratio (Fig 2). The software works through the following algorithms: (1) selection of color component; (2) artifact rejection (bubbles and re£ections); (3) determination of threshold; (4) thresholding; (5) labelingde¢nition of hair regions; (6) deselecting of small regions (smaller than minimal hair length); (7) tattoo elimination (works by using the fact that the tattoo is a large, dark region located in the center of the image); (8) analysis of each hair region: (a) search for the longest straight line (ful¢lling several prede¢ned conditions) at the edge of the analyzed hair region, and (b) reduction of hair region of detected hair; (9) repetition of steps 8a and 8b until no more hair is found; (10) repetition of analysis of all hair regions; and (11) calculation of number of hairs, hair density, and mean/median hair thickness/sum of hair thickness. The software was validated using more than 500 images, which were taken from study participants.
RESULTS Total time ‘‘hands-on’’ for TrichoScan operator Each complete procedure required 15^20 min. The total time ‘‘handson’’ for the TrichoScan operator was approximately 8^12 min (Figs 1A^J and 2). E¡ect of the hair dye In preliminary experiments we tried to analyze fair or gray hair with the TrichoScan software; however, these hairs produced only little contrast, whereas coloring the hairs resulted in a marked increase in hair detectability and did not interfere with the four parameters of hair growth. The dye must be applied for 11^13 min. More than 13 min would unintentionally dye the scalp skin. Fewer than 11 min resulted in incomplete staining. E¡ect of the tattoo In the experiments presented we used a single black tattoo. During the analysis it became clear that black ink interfered with the detectability of the stained hairs. Therefore, in future studies we will use red ink for the tattoo.
Figure 1. Stepwise illustration of the complete TrichoScan procedure. (A) A representative area of the scalp is chosen and the plastic template is applied; (B) all hairs are carefully combed through the plastic template; (C) the hairs are shaved on the scalp surface; (D) the shaved area is 1.8 cm2 in size; (E) 1 cm of dye is applied on to a wood stick; (F) three drops of developer are mixed (G) with the dye; (H) the dye is carefully applied on to the shaved area; (I) after 12 min the dye is carefully removed with an alcoholic solution; (J) digital images are taken at 20- and 40-fold magni¢cation while the area is still wet.
Precision and sensitivity The algorithm excludes all air bubbles, dust, small hemangiomas, nevi, scales, etc. from the calculation without interfering with the number of detectable hairs. In doing so, only hairs are counted and the precision of the method therefore approximates 100%. The detection limit of the software is 5 mm in thickness. Hairs smaller than 5 mm can therefore not be analyzed. Measurement of hair thickness and hair number/analysis of intraoperator error In 10 volunteers the hair number (Fig 3) and cumulative hair thickness (Fig 4) was analyzed in the same area, three times by the same investigator. The percentage of variation in hair counts between volunteers, the intraclass correlation, was estimated at 90.9%. The intraclass correlation for cumulative hair thickness was 90.6%.
VOL. 8, NO. 1 JUNE 2003
TRICHOSCAN FOR HAIR MEASUREMENT
111
Figure 2. Example of the TrichoScan analysis. The ¢gure illustrates a digital image taken at 20-fold magni¢cation (left side of the image) and shows the area of 0.65 cm2 (blue circle), which is analyzed with the TrichoScan software. The TrichoScan results are illustrated on the right side, where the detected hairs are illustrated with di¡erent colors. Red hairs are nongrowing hairs (telogen), green hairs are growing hairs (anagen), and yellow hairs touch the borders of the circle. The right lower part of the ¢gure shows a histogram of the di¡erent hair lengths detected by the TrichoScan software. Furthermore, the cumulative hair thickness and number of terminal and vellus hairs is calculated.
6.0 Cumulative hair thickness (mm)
70
Hair count
60
50
40
30
5.0
4.0
3.0
2.0
20 Eh Fd
If
Kj Ms Rb Rh Sf Th Uw
Subject
Eh Fd
If
Kj Ms Rb Rh Sf Th Uw Subject
Figure 3. Analysis of intraclass correlation. The intraclass correlation of three di¡erent measurements in 10 volunteers (subjects) from the same investigator is shown for hair counts.
Figure 4. Analysis of intraclass correlation. The intraclass correlation of three di¡erent measurements in 10 volunteers (subjects) from the same investigator is shown for cumulative hair thickness.
Analysis of interoperator error In ¢ve volunteers, hair number (Fig 5) and cumulative hair thickness (Fig 6) was analyzed in the same area once, but by two independent
investigators. The interclass correlation was estimated for hair counts at 97.6% and for the cumulative hair thickness at 96.4%.
HOFFMANN
JID SYMPOSIUM PRECEEDINGS
Analysis of total hair counts and cumulative hair thickness in volunteers without AGA, with untreated AGA, and in AGA treated with ¢nasteride In individuals a¡ected by AGA, a transitional area of hair loss between normal hair and the balding area was de¢ned and area of 0.225 cm2 was analyzed at 40-fold magni¢cation. Twelve men were treated with ¢nasteride, whereas ¢ve men with AGA remained untreated. For both variables di¡erences between results after 3 mo (6 mo) and baseline were calculated. These di¡erences were analyzed using a one-sample t test. In controls and untreated men we noticed no signi¢cant di¡erence in the number of hairs within the observation time of 6 mo. By contrast, men treated with ¢nasteride showed a continuous increase at 3 mo ( þ 17%; p ¼ 0.055) and at 6 mo ( þ 20%; p ¼ 0.021) in the number of hairs within the analyzed area (Fig 7a,b and Table I). In 11 volunteers without AGA (controls) we observed no signi¢cant di¡erence in the cumulative hair thickness within the observation time of 6 mo (Fig 8a,b), whereas untreated men showed a continuous and signi¢cant decrease in the overall thickness of hairs 3 mo (^ 5%) and 6 mo (^ 5%) after the initial visit (Fig 8a,b, Table I). By contrast, men treated with ¢nasteride showed a continuous and signi¢cant increase in hair thickness within the analyzed area (Table I) after 3 mo ( þ 11%: p ¼ 0.034) and 6 mo ( þ 18%; p ¼ 0.006) (Fig 8a,b), compared with values obtained at baseline. Analysis of anagen/telogen ratios and hair growth rates at the vertex and the occiput in volunteers with AGA The variables analyzed were the percentage of anagen hairs (growing hairs) and the hair growth rate (di¡erence of the length of anagen hairs minus the length of telogen hairs divided by the time of measurement after clipping). (The telogen hairs are de¢ned as nongrowing hairs. Telogen and catagen hairs cannot be di¡erentiated.) For both variables an ANOVA was calculated with the ¢xed factors diagnosis (AGA/control) and investigator (investigator 1/investigator 2) and the random factor location nested under the diagnosis. The p-values were given for a two-sided problem. Figure 9 shows the original values of the percentage of anagen hairs for the di¡erent diagnoses and investigators with 95% con¢dence intervals of the means. The same is shown (Fig 10) for the length di¡erences between anagen and telogen hairs. The model of the percentage of anagen hairs explains 96% of the variance, the hair growth 85%. Both models are highly signi¢cant. Table II shows the model and the single e¡ects. For both variables the diagnosis is
70
Hair count
60
50
40
30
20 Cm
Eh
Kh
Th
Uh
Subject
Figure 5. Analysis of intraclass correlation. The intraclass correlation of one measurement in ¢ve volunteers (subjects) from two di¡erent investigators is shown for hair counts and for cumulative hair thickness (Fig 6).
6.0 Cumulative hair thickness (mm)
112
5.0
4.0
3.0
2.0 Cm
Eh
Kh
Th
Uh
Subject
Figure 6. Analysis of intraclass correlation. The intraclass correlation of one measurement in ¢ve volunteers (subjects) from two di¡erent investigators is shown for hair counts (Fig 5) and for cumulative hair thickness.
highly signi¢cant. These results show that an AGA-a¡ected scalp has a lower number of anagen hair follicles (Fig 9) and that these hair follicles grow slower (Fig 10), compared with hair follicles at the occiput. DISCUSSION
Numerous hair diseases such as the scarring alopecias, alopecia areata, or trichotillomania, usually do not need a quantitative method to evaluate the amount of hair shedding. Androgenetic e¥uvium, however, the most common form of hair loss, is typically di⁄cult to quantify and at present simple but reliable o⁄ce analytical procedures have not been developed. Although scalp biopsies can be justi¢ed in that microscopic examination of scalp skin a¡ected by AGA can identify and quantify any changes resulting from treatment, this invasive technique is often not suitable to monitor patients over a prolonged period of time. The classical trichogram is harmless and easy to use, but not reliable. AGA can be de¢ned as an androgen-dependent process in genetically predisposed individuals, where balding is due to continuous miniaturization of a¡ected hair follicles, changing large terminal hair follicles into small vellus-like hairs (Whiting et al, 1999; Ho¡mann and Happle, 2000a,b). Any successful treatment should therefore stop or reverse the process of hair follicle miniaturization, increase the number of terminal hair follicles, while reducing vellus hair counts (Whiting et al, 1999) or increase the number of anagen hairs (Van Neste et al, 2000). This concept is illustrated by the phase III studies of men with AGA treated with ¢nasteride (Kaufman et al, 1998). In these studies, macrophotographs were taken and hairs were counted. This technique produces counts of ‘‘visible’’ hairs, which means that tiny vellus-like hairs cannot be seen or counted. During treatment, however, these vellus-like hair follicles enlarge and subsequently increase the hair count results when the macrophotograph method is used. A major disadvantage of this technique is that it cannot monitor the expected continuous increase in hair thickness during treatment. As a consequence the phase III studies of men with AGA treated with ¢nasteride revealed that the increase in hair counts reaches a plateau after 1 y of treatment, whereas the hair coverage analyzed by global photographs increases continuously (Kaufman et al, 1998). This increase in hair coverage is due to an increase in hair thickness as shown by histologic examination (Whiting et al, 1999), the direct measurement of hair thickness (Steiner et al, 2000), and by a continuous increase in hair weight (Whiting, 2001). Although the Ludwig pattern of AGA in women di¡ers
VOL. 8, NO. 1 JUNE 2003
Table I. Trichoscan results from treated (¢nasteride) untreated men and controls 3 and 6 months after baseline
20
Variable
10
Group
Hair count
Control
0
AGA untreated AGA treated
-10 Diff. after 3 months
Diff. after 6 months
-20 11
11
5
control
Change in hair count
B
5
12
Control
12
AGA without
AGA with
treatment
treatment
AGA untreated AGA treated
140%
Time
df
t
p
After 3 mo After 6 mo After 3 mo After 6 mo After 3 mo After 6 mo After 3 mo
10 10 4 4 11 11 10
0.265 0.029 0.199 0.106 2.149 2.688 0.012
0.796 0.977 0.852 0.921 0.055 0.021 0.991
After 6 mo After 3 mo After 6 mo After 3 mo After 6 mo
10 4 4 11 11
0.540 0.739 1.161 2.414 3.433
0.601 0.501 0.310 0.034 0.006
120%
A
100%
80% Diff. after 3 months
60%
Diff. after 6 months N=
11
11
control
5
5
12
12
AGA without
AGA with
treatment
treatment
Figure 7. (a) Hair counts was analyzed for 6 mo in 11 volunteers without AGA, in ¢ve untreated men with AGA, and in 12 men treated with ¢nasteride (1 mg per day). In controls and untreated men we noticed no significant di¡erence in the number of hairs within the observation time of 6 mo (between the values at baseline and after 6 mo). In contrast those men treated with ¢nasteride showed a continuous increase (mean with 95% con¢dence interval) at 3 mo (p ¼ 0.055) and at 6 mo (p ¼ 0.021) in the number of hairs within the analyzed area compared with the values at baseline. (b) Hair counts was analyzed for 6 mo in 11 volunteers without AGA, in ¢ve untreated men with AGA, and in 12 men treated with ¢nasteride (1 mg per day). Quotients of hair count between measurements after 3 and 6 mo and ¢rst measurement for healthy subjects (controls), treated, and untreated patients with AGA (mean with 95% con¢dence interval).
in appearance (Ludwig, 1977) from the Hamilton pattern occurring in men, the pathophysiologic mechanisms seem to be the same, because female AGA patients treated with cyproterone acetate (Mortimer et al, 1984; Peereboom-Wynia et al, 1989; Dawber, 1995) or minoxidil (Price and Menefee, 1990; Rushton, 1993) experience an increase in hair thickness (Mortimer et al, 1984; Vanderveen et al, 1984; Peereboom-Wynia et al, 1989; Dawber, 1995) or hair weight (Price and Menefee, 1990). Therefore, a reliable hair counting method should primarily be able to calculate the number and thickness of hairs, which is stable within at least 1 cm above the scalp (Jackson et al, 1972; Hutchinson and Thompson, 1997), in a de¢ned area of the scalp. From a clinical perspective, hair thickness is very important (de Lacharriere et al, 2001), whereas the growth rate (mm per day) and the anagen/telogen ratio are of secondary importance. As early as 1964, Barman et al (1964) related a method that used optical contact microscopy to calculate these parameters, and much later Hayashi et al (1991) and recently D’Amico et al (2001) described a similar approach for the measurement of hair growth using optical microscopy and computer analysis; however, these investigators were unable to automate the process of calculation and measured the thickness of hairs visually with the cursor on the computer monitor. They calculated that results from di¡erent
Change in cumulative hair thickness (mm)
N=
Cumulative hair thickness
2.0
1.0
0.0
-1.0 Diff. after 3 months Diff. after 6 months
-2.0 N=
11
11
5
control
B Change in cumulative hair thickness
Change in hair count (95% CI)
A
113
TRICHOSCAN FOR HAIR MEASUREMENT
5
12
AGA without treatment
12
AGA with treatment
140%
120%
100%
80%
Diff. after 3 months
Diff. after 6 months
60% N = 11
11 control
5
5
AGA without treatment
12
12
AGA with treatment
Figure 8. (a) Cumulative hair thickness were analyzed for 6 mo in 11 volunteers without AGA, in ¢ve untreated men with AGA, and in 12 men treated with ¢nasteride (1 mg per day). Untreated men showed a continuous and signi¢cant decrease in the overall thickness of hairs 3 and 6 mo after the initial visit (baseline). In contrast those men treated with ¢nasteride showed in comparison with the baseline visit a continuous and signi¢cant increase in the hair thickness within the analyzed area after 3 mo (p ¼ 0.034) and 6 mo (p ¼ 0.006). (b) Hair counts and cumulative hair thickness was analyzed for 6 mo in 11 volunteers without AGA, in ¢ve untreated men with AGA, and in 12 men treated with ¢nasteride (1 mg per day). Quotients of cumulative hair thickness between measurements after 3 and 6 mo and ¢rst measurement for healthy subjects (controls), treated, and untreated patients with AGA (mean with 95% con¢dence interval).
investigators, but from the same image, di¡er by 7 8.4%, which makes such a semiautomatic method unsuitable for clinical practice. A similar approach has been tested using a phototrichogram,
114
HOFFMANN
JID SYMPOSIUM PRECEEDINGS
Table II. Trichoscan results for anagen hair and growth rate in AGA-a¡ected and una¡ected scalp
Anagen Hairs [%] (95% CI)
100
Anagen hairs 80
Hair growth 70
AGA
60 N=
df
F ratio
p
Model Diagnosis Investigator Image (diagnosis) Model Diagnosis Investigator Image (diagnosis)
34 1 1 31 34 1 1 31
17.90 57.60 0.15 7.18 4.95 15.10 2.00 3.40
o0.0001 o0.0001 0.702 o0.0001 o0.0001 o0.001 0.168 o0.001
control 18
14
18
A
13
B
Investigator
Figure 9. Analysis of anagen/telogen ratio. While de¢ning anagen hairs as growing and telogen hairs as nongrowing hairs 3 d after shaving, the TrichoScan is able to calculate a digital trichogram. This ¢gure illustrates the results (mean with 95% con¢dence interval) from two investigators, who analyzed in 18 volunteers with AGA the proportion of anagen hairs at the vertex and at the occiput (only 14 images). Compared with the occiput, the AGA-a¡ected scalp reveals a decreased number of anagen hairs. Both investigators produced similar results.
.42
Hair Growth [mm/day] (95% CI)
E¡ect 90
.40 .38 .36 .34 AGA
.32
control
.30 N=
18
14
18
A
13
B
Investigator
Figure 10. Analysis of hair growth rate. This ¢gure illustrates the results (mean with 95% con¢dence interval) from two investigators, who analyzed in 18 volunteers with AGA the and hair growth rate at the vertex and at the occiput (only 14 images). Compared with the occiput, the AGAa¡ected scalp reveals slower growing anagen hairs. Both investigators produced similar results.
which has proven to be a suitable, noninvasive tool to monitor hair growth phases in situ. This technique has been improved with image analysis (Van Neste et al, 1989) and later using immersion oil and digital contrast enhancement (Van Neste et al, 1992). Despite markedly improved images and more accurate quantitative data (Van Neste et al, 1992), however, automated analysis was not possible. Until now, analysis of images has been tedious and time consuming. Unsuccessful attempts to automate the process have been reported several times (Pel¢ni and Calligaro, 1986; Hayashi et al, 1991; Van Neste et al, 1992). The major problem has been that scalp hair follicles grow in groups (follicular units) rather than singly, and therefore neighboring hair follicles typically overlap or may be aligned in parallel. Furthermore, photographic analysis software requires good contrast between the hair follicles and the scalp skin, and the fact that many hairs lose their natural pigmentation due to aging or AGA, makes them much more di⁄cult to
detect. We have overcome this di⁄culty by coloring the hairs prior to taking the images, without any negative e¡ect on the collected data. Furthermore, we have created automatic software for the analysis of the aforementioned parameters of hair growth. Because this described technique is a modi¢ed and computerized trichogram we called it TrichoScan. Images were taken with a video system using ELM. ELM is a standard procedure used in analyzing melanocytic nevi (Steiner et al, 1987; Braun-Falco et al, 1990; Kreusch et al, 1992; Krischer et al, 1999), and many dermatologists in Europe already use ELM systems in daily clinical practice. These devices produce high-quality and reproducible digital images, because the images are always taken at the same distance from the skin surface. Our results suggest that ELM systems can be used to evaluate patients complaining of androgenetic e¥uvium and for monitoring responses to therapy. Variations that normally occur in hair length, weight, thickness, etc. can be assayed either with respect to standardized values, or by comparing measurements made on two or more occasions over a given period of time. Similar means of assay must be employed to study changes in hair growth that may occur with regard to age or illness. The margin of error of the techniques and the instruments employed should be smaller than the magnitude of the variations to be measured. As our results show, the TrichoScan ful¢lls these criteria and has advantages over standard procedures used thus far for hair measurements. 1 It is investigator independent. In other studies using the unit area trichogram, substantial di¡erences between investigators were noted. In these studies a signi¢cantly larger mean total hair counts were reported from experienced compared with inexperienced observers (Rushton et al, 1989). Our results show that this is not the case for the TrichoScan technique. 2 Many methods are not strictly validated. The hair weight test is a good example where the hair is clipped in a de¢ned target area; however, the sample error for di¡erent investigators is unknown. This is mainly due to the methodology itself, because once hairs are clipped a second investigator cannot clip the same area again to assess the reproducibility of the method. In contrast, the TrichoScan is highly validated with de¢ned values for intraclass correlation between the same and di¡erent investigators. This is of crucial importance for clinical studies. In cases where the expected di¡erences between placebo and actively treated patients is known, the minimum number of patients necessary to prove this di¡erence can be calculated. 3 Some methods are associated with considerable discomfort to the patient such as the repeated plucking of hairs required by the trichogram technique. The TrichoScan relies on a small analyzed area of the scalp, which is afterwards barely visible. The tiny tattoo is the only discomfort patients will recognize. 4 Some methods to count hairs are tedious and time-consuming. By contrast, TrichoScan can be used by an experienced operator within 8^12 min ‘‘hands on’’. 5 The amount of equipment is small. Many dermatologists already have ELM systems and these physicians would only need the TrichoScan software.
VOL. 8, NO. 1 JUNE 2003
TRICHOSCAN FOR HAIR MEASUREMENT
6 Many methods only count pigmented hair shafts.Via staining the hairs prior the analysis, the TrichoScan is able to analyze fair and gray hairs as well. 7 Most techniques require a rather long observation time to be able to detect signi¢cant di¡erences between groups of patients. In contrast, the TrichoScan allows the detection of a statistically signi¢cant increase in hair thickness even after 3 mo of treatment in a rather small cohort of patients. Therefore, the overall advantage of the TrichoScan is that this technique can be used for clinical studies to compare placebo vs treatment or to compare di¡erent capacities of di¡erent hair growth promoting substances. This technique can be used to study AGA or other forms of di¡use hair loss, Moreover, it can be adopted to study the e¡ect of drugs and laser treatment on hypertrichosis and hirsutism. The work of U. Ellwanger and H. Lˇdtke (Datenanalyze und Angewandte Informatik GbR [www.datinf.de], Brunnenstr. 14, 72074 Tˇbingen, Germany) in programming the software is greatfully appreciated.
REFERENCES Barman JM, Pecoraro V, Astore I: Method, technic and computations in the study of the trophic state of human scalp hair. J Invest Dermatol 42:421^425, 1964 Barth JH, Rushton DH: Measurement of hair growth. In: Serudp J, Jemee GBE (eds). Non-Invasive Methods and the Skin, Vol. 1. Ann Arbor: CRP Press, 1995: pp 543^548 Blume-Peytavi U, Orfanos CE: Microscopy of the hair. In: Serudp J, Jemee GBE (eds). Non-Invasive Methods and the Skin, Vol. 1. Ann Arbor: CRP Press, 1995: pp 549^554 Bouhanna P: The phototrichogram, a macrophotographic study of the scalp. Bioengineer Skin 3:265, 1985 Braun-Falco O, Stolz W, Bilek P, Merkle T, Landthaler M: The dermatoscope: Hautarzt. 41:131^136, 1990 Can¢eld D: Photographic documentation of hair growth in androgenetic alopecia. Dermatol Clin 14:713^721, 1996 D’Amico D, Vaccaro M, Guarneri F, Borgia F, Cannavo S, Guarneri B: Phototrichogram using videomicroscopy: A useful technique in the evaluation of scalp hair. Eur J Dermatol 11:17^20, 2001 Dawber RPR: Haarwachstumsdiagnostik und das Haarwachstum stimulierende Substanzen. In: Tebbe B, Goerdt S, Orfanos CE (eds). Dermatologie, Heutiger Stand. Stuttgart: Georg Thieme Verlag, 1995:pp 243^248 Friedel J, Will F, Grosshans E: Phototrichogram. Adaptation, standardization and applications. Ann Dermatol Venereol 116:629^636, 1989 Guarrera M, Ciulla MP: A quantitative evaluation of hair loss: The phototrichogram. J Appl Cosmetol 4:61^66, 1986 Hamilton JB: Patterned loss of hair in man: Types and incidence. Ann N Y Acad Sci 53:708^728, 1951 Hayashi S, Miyamoto I, Takeda K: Measurement of human hair growth by optical microscopy and image analysis. Br J Dermatol 125:123^129, 1991 Headington JT: Transverse microscopic anatomy of the human scalp. A basis for a morphometric approach to disorders of the hair follicle. Arch Dermatol 120:449^456, 1984 Ho¡mann R, Happle R: Current understanding of androgenetic alopecia. Part I: Etiopathogenesis. Eur J Dermatol 10:319^327, 2000a Ho¡mann R, Happle R: Current understanding of androgenetic alopecia. Part II: Clinical aspects and treatment. Eur J Dermatol 10:410^417, 2000b Hutchinson PE,Thompson JR: The cross-sectional size and shape of human terminal scalp hair. Br J Dermatol 136:159^165, 1997
115
Jackson D, Church RE, Ebling FJ: Hair diameter in female baldness. Br J Dermatol 87:361^367, 1972 Kaufman KD, Olsen EA, Whiting D, et al: Finasteride in the treatment of men with androgenetic alopecia. Finasteride Male Pattern Hair Loss Study Group. J Am Acad Dermatol 39:578^589, 1998 Kreusch J, Rassner G, Trahn C, Pietsch-Breitfeld B, Henke D, Selbmann HK: Epiluminescent microscopy: A score of morphological features to identify malignant melanoma. Pigment Cell Res Suppl 2:295^298, 1992 Krischer J, Braun RP, Toutous-Trellu L, Saurat JH, Pechere M: Kaposi’s sarcoma: A new approach of lesional follow-up using epiluminescent light microscopy. Dermatology 198:420^422, 1999 de Lacharriere O, Deloche C, Misciali C, et al: Hair diameter diversity: A clinical sign re£ecting the follicle miniaturization. Arch Dermatol 137:641^646, 2001 Ludwig E: Classi¢cation of the types of androgenetic alopecia (common baldness) occurring in the female sex. Br J Dermatol 97:247^254, 1977 Maguire HC, Kligman AM: Hair plucking as a diagnostic tool. J Invest Dermatol 43:77^79, 1964 Mortimer CH, Rushton H, James KC: E¡ective medical treatment of common baldness in women. Clin Exp Dermatol 9:342^350, 1984 Pecoraro V, Astore I, Barman JM: Growth rate and hair density of the human axilla. A. Comparative study of normal males and females and pregnant and postpartum females. J Invest Dermatol 56:362^365, 1971 Peereboom-Wynia JD, van der Willigen AH, van Joost T, Stolz E: The e¡ect of cyproterone acetate on hair roots and hair shaft diameter in androgenetic alopecia in females. Acta Derm Venereol 69:395^398, 1989 Pel¢ni C, Calligaro A: Evaluation of hair growth by means of morphometric computerized analysis. J Appl Cosmetol 4:67^76, 1986 Price VH, Menefee E: Quantitative estimation of hair growth. I. Androgenetic alopecia in women: E¡ect of minoxidil. J Invest Dermatol 95:683^687, 1990 Rushton DH: Management of hair loss in women. Dermatol Clin 11:47^53, 1993 Rushton H, James KC, Mortimer CH: The unit area trichogram in the assessment of androgen-dependent alopecia. Br J Dermatol 109:429^437, 1983 Rushton DH, Unger WP, Cotterill PC, Kingsley P, James KC: Quantitative assessment of 2% topical minoxidil in the treatment of male pattern baldness. Clin Exp Dermatol 14:40^46, 1989 Rushton DH, de Brouwer B, de Coster W, van Neste DJ: Comparative evaluation of scalp hair by phototrichogram and unit area trichogram analysis within the same subjects. Acta Derm Venereol 73:150^153, 1993 Saitoh M, Uzuka M, Sakamoto M: Human hair cycle. J Invest Dermatol 54:65^81, 1970 Steiner A, Pehamberger H, Wol¡ K: Improvement of the diagnostic accuracy in pigmented skin lesions by epiluminescent light microscopy. Anticancer Res 7:433^ 434, 1987 Steiner D, Bedin V, Pasello RR: Hair shaft diameter evaluation in AGA before and after ¢nasteride 1 mg per day. Poster Nr 284. Annual Meeting of the AAD 2000 Van Neste D: Human scalp hair growth and loss evaluation methods: Is there simple and reliable method? Exp Dermatol 8:299^301, 1999 Van Neste DJ: Contrast enhanced phototrichogram (CE-PTG): An improved noninvasive technique for measurement of scalp hair dynamics in androgenetic alopeciavalidation study with histology after transverse sectioning of scalp biopsies. Eur J Dermatol 11:326^331, 2001 Van Neste DJJ, Dumrotier M, De Coster W: Phototrichogram analysis. Technical aspects and problems in relation with automated quantitative evaluation of hair growth by computer-assisted image analysis. In: Van Neste DJJ, Lachapelle JM, Antoine JL (eds). Trends in Human Hair Growth and Alopecia Research. Dordrecht: Kluwer, 1989:pp 155^165 Van Neste DJJ, Dumrtier M, de Brouwer B, De Coster W: Scalp immersion proxigraphy (SIP): An improved imaging technique for phototrichogram analysis. J Eur Acad Dermatol Venereol 1:187^191, 1992 Van Neste DJ, de Brouwer B, De Coster W: The phototrichogram: Analysis of some technical factors of variation. Skin Pharmacol 7:67^72, 1994 Van Neste D, Fuh V, Sanchez-Pedreno P, et al: Finasteride increases anagen hair in men with androgenetic alopecia. Br J Dermatol 143:804^810, 2000 Vanderveen EE, Ellis CN, Kang S, et al: Topical minoxidil for hair regrowth. J Am Acad Dermatol 11:416^421, 1984 Whiting DA: Advances in the treatment of male androgenetic alopecia: A brief review of ¢nasteride studies. Eur J Dermatol 11:332^334, 2001 Whiting DA, Waldstreicher J, Sanchez M, Kaufman KD: Measuring reversal of hair miniaturization in androgenetic alopecia by follicular counts in horizontal sections of serial scalp biopsies: Results of ¢nasteride 1 mg treatment of men and postmenopausal women. J Invest Dermatol Symp Proc 4:282^284, 1999
TrichoScan: A Novel Tool For The Analysis of Hair Growth In Vivo Rolf Ho¡mann Department of Dermatology, Philipp University, Marburg, Germany
Hair loss or hair thinning is a common complaint in clinical dermatology, and patients seeking advice for hair loss are not necessarily bald. Because the e¡ects of treatment attempts are hard to measure, there is need for a sensitive tool to monitor hair loss and treatment responses. Such a method must be able to analyze the biologic parameters of hair growth, which are: (i) hair density (n per cm2); (ii) hair diameter (lm); (iii) hair growth rate (mm per day); and (iv) anagen/telogen ratio. Herein we present the TrichoScan as a method that combines epiluminescence microscopy with automatic digital image analysis for the measurement of human, and potentially animal hair, in situ. The TrichoScan is able to analyze all four parameters of hair growth with a so-called intraclass correlation of approximately 91% within the same TrichoScan operator and an intraclass
correlation of approximately 97% for di¡erent TrichoScan operators. The application of the technique is demonstrated by comparing the hair parameters in individuals without apparent hair loss, men with untreated androgenetic alopecia, and men after treatment with ¢nasteride (1 mg per day). We were able to detect a signi¢cant increase in hair counts and cumulative hair thickness 3 and 6 mo after treatment. Advantages of the TrichoScan are that it can be used for clinical studies to compare placebo versus treatment, to compare di¡erent capacities of hair growth promoting substances, to study androgenetic alopecia and other forms of di¡use hair loss, and to study the e¡ects of drugs and laser treatment on hypertrichosis and hirsutism. Key words: alopecia/computer analysis/hair/measurement/treatment. JID Symposium Proceedings 8:109 ^115, 2003
H
(Rushton et al, 1993; Van Neste, 1999). Therefore, an operatorfriendly and patient-friendly, inexpensive, validated, and reliable method is a rational need. A useful method must be able to analyze the biologic parameters of hair growth, which are: (i) hair density (n per cm2); (ii) hair diameter (mm); (iii) hair growth rate (mm per day); and (iv) anagen/telogen ratio. This paper describes the TrichoScan as such a method, which combines standard epiluminescence microscopy (ELM) with automatic digital image analysis for the measurement of human, and potentially animal hair, in situ. Application of the technique is demonstrated by comparing the four hair parameters in individuals without apparent hair loss, men with untreated AGA, and men after treatment with ¢nasteride (1 mg per day).
air loss or hair thinning is a common complaint in clinical dermatology, and patients seeking advice for hair loss are not necessarily bald. In established cases of androgenetic alopecia (AGA) characteristic patterns are easily discernible; however, especially in women, the clinician is often challenged by patients with initial stages of AGA where hair loss is reported but alopecia is not recognizable or the e¡ects of treatment attempts are hard to measure. Consequently, there is a need for a sensitive tool to monitor hair loss and treatment responses. Numerous methods have been reported (Barth and Rushton, 1995) to assess rates of hair growth. These techniques can be classi¢ed as either (i) invasive, such as biopsies (Headington, 1984; Whiting et al, 1999); (ii) semi-invasive, such as the trichogram (Maguire and Kligman, 1964; Blume-Peytavi and Orfanos, 1995) and unit area trichogram (Rushton et al, 1983); or (iii) noninvasive, such as global hair counts (Can¢eld, 1996) and the phototrichogram (Saitoh et al, 1970; Bouhanna, 1985; Guarrera and Ciulla, 1986; Friedel et al, 1989; Van Neste et al, 1989, 1994). Quantitative methods to analyze human hair growth and hair loss are necessary to determine the e⁄cacy of hair promoting drugs. In reviewing capabilities of the di¡erent methods, a common theme emerges, that most techniques are of little use to the clinician because they are time consuming, costly, and/or di⁄cult to perform
MATERIALS AND METHODS Volunteers and patients A total of 56 persons (25 women, age range 25^48 y, mean 34 y; 31 men, age range 26^39 y, mean 32 y) were studied. Ten of 56 volunteers (¢ve women, age range 25^48 y, mean 34 y; ¢ve men, age range 26^39 y, mean 32 y) were recruited for the initial experiments to analyze the reproducibility of the method. The measured parameters were hair thickness and hair numbers at the occiput. Seventeen male patients (age range 25^48 y, mean 34 y), who had recognized progressive thinning of hair and hair loss for more than 4 y were included to analyze the progression of AGA, with and without treatment. All patients presented the clinical ¢nding of mild to moderate AGA with various degrees of involvement that were classi¢ed according to the Hamilton scale (II^V) (Hamilton, 1951). Patients with other forms of alopecia were excluded. Twelve of the 17 male patients with AGA were treated with Propecias (Merch & Co., Inc. Whitehouse Station, Rahway, NJ) (1 mg ¢nasteride per day) for 6 mo. All patients actively treated had
Manuscript accepted for publication February 1, 2003 Reprint requests to: Rolf Ho¡mann MD, Department of Dermatology, Philipp University, Deutschhausstrasse 9, 35033 Marburg, Germany. Email: [email protected]. Abbreviations: AGA, androgenetic alopecia; ELM, epiluminescence microscopy.
0022-202X/03/$15.00 . Copyright r 2003 by The Society for Investigative Dermatology, Inc. 109
110
HOFFMANN
had no treatment whatsoever for hair loss for at least 1 mo before initiating the study. Eleven of 56 healthy male volunteers (age range 28^55 y, mean 36 y) who had experienced no episodes of hair thinning or hair loss, recent illnesses, or general health disturbances, were recruited as a control group. Clinical examination revealed no evidence of any hair disorder with either the female or the male volunteers. The measured parameters were hair thickness and hair numbers at the vertex. For the analysis of daily hair growth and the anagen/telogen ratio, an additional 18 volunteer patients with AGA were recruited.
Clipping of hairs In individuals a¡ected by AGA, a transitional area of hair loss between normal hair and the balding area was de¢ned and an area of 1.8 cm2 was clipped (Hairliner, Wella Germany Darmstadt, Germany) (Fig 1A^D). In volunteers without AGA (controls) the vertex was chosen for clipping. All clipped areas were landmarked with a central, single black tattoo. The tattoo was visible throughout the study. In the 18 volunteers who were recruited for the analysis of the anagen/telogen ratios, the scalp was clipped at two locations (vertex and occiput) and was analyzed by two investigators with the TrichoScan software.
JID SYMPOSIUM PRECEEDINGS
ELM of clipped hairs Gray or fair hairs have only limited contrast in comparison with the scalp (Van Neste, 2001). Therefore, the clipped hairs within the target area were dyed for 12 min (Fig 1E^H) with a commercially available solution (RefectoCils, Gschwentner, Vienna, Austria), which is normally used to color eye brows and lashes. The approach of dying hairs for hair growth studies has been reported to give the same results as uncolored hairs (Pecoraro et al, 1971). For the analysis of hair number and thickness, hairs were colored immediately after shaving and for the analysis of the hair growth rate and anagen/telogen ratio, hairs were colored 3 d after shaving. Thereafter, the colored area was cleaned (Fig 1I) with an alcoholic solution (Kodans Spray, Schˇlke & Mayr, Vienna, Austria) and digital images were obtained at 20-fold (analyzed area: 0.62 cm2) and 40-fold (analyzed area: 0.225 cm2) magni¢cation by means of a digital ELM system (Foto¢nder DERMA, Teachscreen Software, Bad Birnbach, Germany) while the area was still wet (Fig 1J). This digital camera was equipped with a rigid contact lens that ensured images were always taken at the same distance from the scalp. Because the camera must be pressed on to the scalp, the hairs were always £attened. Images were taken at day 0, immediately after clipping, 2 and 3 d after clipping, and 3 and 6 mo after the initial visit, respectively. Two di¡erent investigators each took three images from the same patient at every visit. Software for digital image and statistical analysis Software was developed to analyze (TrichoScan) hair density (n per cm2), hair diameter (mm), hair growth rate (mm per day), and anagen/telogen ratio (Fig 2). The software works through the following algorithms: (1) selection of color component; (2) artifact rejection (bubbles and re£ections); (3) determination of threshold; (4) thresholding; (5) labelingde¢nition of hair regions; (6) deselecting of small regions (smaller than minimal hair length); (7) tattoo elimination (works by using the fact that the tattoo is a large, dark region located in the center of the image); (8) analysis of each hair region: (a) search for the longest straight line (ful¢lling several prede¢ned conditions) at the edge of the analyzed hair region, and (b) reduction of hair region of detected hair; (9) repetition of steps 8a and 8b until no more hair is found; (10) repetition of analysis of all hair regions; and (11) calculation of number of hairs, hair density, and mean/median hair thickness/sum of hair thickness. The software was validated using more than 500 images, which were taken from study participants.
RESULTS Total time ‘‘hands-on’’ for TrichoScan operator Each complete procedure required 15^20 min. The total time ‘‘handson’’ for the TrichoScan operator was approximately 8^12 min (Figs 1A^J and 2). E¡ect of the hair dye In preliminary experiments we tried to analyze fair or gray hair with the TrichoScan software; however, these hairs produced only little contrast, whereas coloring the hairs resulted in a marked increase in hair detectability and did not interfere with the four parameters of hair growth. The dye must be applied for 11^13 min. More than 13 min would unintentionally dye the scalp skin. Fewer than 11 min resulted in incomplete staining. E¡ect of the tattoo In the experiments presented we used a single black tattoo. During the analysis it became clear that black ink interfered with the detectability of the stained hairs. Therefore, in future studies we will use red ink for the tattoo.
Figure 1. Stepwise illustration of the complete TrichoScan procedure. (A) A representative area of the scalp is chosen and the plastic template is applied; (B) all hairs are carefully combed through the plastic template; (C) the hairs are shaved on the scalp surface; (D) the shaved area is 1.8 cm2 in size; (E) 1 cm of dye is applied on to a wood stick; (F) three drops of developer are mixed (G) with the dye; (H) the dye is carefully applied on to the shaved area; (I) after 12 min the dye is carefully removed with an alcoholic solution; (J) digital images are taken at 20- and 40-fold magni¢cation while the area is still wet.
Precision and sensitivity The algorithm excludes all air bubbles, dust, small hemangiomas, nevi, scales, etc. from the calculation without interfering with the number of detectable hairs. In doing so, only hairs are counted and the precision of the method therefore approximates 100%. The detection limit of the software is 5 mm in thickness. Hairs smaller than 5 mm can therefore not be analyzed. Measurement of hair thickness and hair number/analysis of intraoperator error In 10 volunteers the hair number (Fig 3) and cumulative hair thickness (Fig 4) was analyzed in the same area, three times by the same investigator. The percentage of variation in hair counts between volunteers, the intraclass correlation, was estimated at 90.9%. The intraclass correlation for cumulative hair thickness was 90.6%.
VOL. 8, NO. 1 JUNE 2003
TRICHOSCAN FOR HAIR MEASUREMENT
111
Figure 2. Example of the TrichoScan analysis. The ¢gure illustrates a digital image taken at 20-fold magni¢cation (left side of the image) and shows the area of 0.65 cm2 (blue circle), which is analyzed with the TrichoScan software. The TrichoScan results are illustrated on the right side, where the detected hairs are illustrated with di¡erent colors. Red hairs are nongrowing hairs (telogen), green hairs are growing hairs (anagen), and yellow hairs touch the borders of the circle. The right lower part of the ¢gure shows a histogram of the di¡erent hair lengths detected by the TrichoScan software. Furthermore, the cumulative hair thickness and number of terminal and vellus hairs is calculated.
6.0 Cumulative hair thickness (mm)
70
Hair count
60
50
40
30
5.0
4.0
3.0
2.0
20 Eh Fd
If
Kj Ms Rb Rh Sf Th Uw
Subject
Eh Fd
If
Kj Ms Rb Rh Sf Th Uw Subject
Figure 3. Analysis of intraclass correlation. The intraclass correlation of three di¡erent measurements in 10 volunteers (subjects) from the same investigator is shown for hair counts.
Figure 4. Analysis of intraclass correlation. The intraclass correlation of three di¡erent measurements in 10 volunteers (subjects) from the same investigator is shown for cumulative hair thickness.
Analysis of interoperator error In ¢ve volunteers, hair number (Fig 5) and cumulative hair thickness (Fig 6) was analyzed in the same area once, but by two independent
investigators. The interclass correlation was estimated for hair counts at 97.6% and for the cumulative hair thickness at 96.4%.
HOFFMANN
JID SYMPOSIUM PRECEEDINGS
Analysis of total hair counts and cumulative hair thickness in volunteers without AGA, with untreated AGA, and in AGA treated with ¢nasteride In individuals a¡ected by AGA, a transitional area of hair loss between normal hair and the balding area was de¢ned and area of 0.225 cm2 was analyzed at 40-fold magni¢cation. Twelve men were treated with ¢nasteride, whereas ¢ve men with AGA remained untreated. For both variables di¡erences between results after 3 mo (6 mo) and baseline were calculated. These di¡erences were analyzed using a one-sample t test. In controls and untreated men we noticed no signi¢cant di¡erence in the number of hairs within the observation time of 6 mo. By contrast, men treated with ¢nasteride showed a continuous increase at 3 mo ( þ 17%; p ¼ 0.055) and at 6 mo ( þ 20%; p ¼ 0.021) in the number of hairs within the analyzed area (Fig 7a,b and Table I). In 11 volunteers without AGA (controls) we observed no signi¢cant di¡erence in the cumulative hair thickness within the observation time of 6 mo (Fig 8a,b), whereas untreated men showed a continuous and signi¢cant decrease in the overall thickness of hairs 3 mo (^ 5%) and 6 mo (^ 5%) after the initial visit (Fig 8a,b, Table I). By contrast, men treated with ¢nasteride showed a continuous and signi¢cant increase in hair thickness within the analyzed area (Table I) after 3 mo ( þ 11%: p ¼ 0.034) and 6 mo ( þ 18%; p ¼ 0.006) (Fig 8a,b), compared with values obtained at baseline. Analysis of anagen/telogen ratios and hair growth rates at the vertex and the occiput in volunteers with AGA The variables analyzed were the percentage of anagen hairs (growing hairs) and the hair growth rate (di¡erence of the length of anagen hairs minus the length of telogen hairs divided by the time of measurement after clipping). (The telogen hairs are de¢ned as nongrowing hairs. Telogen and catagen hairs cannot be di¡erentiated.) For both variables an ANOVA was calculated with the ¢xed factors diagnosis (AGA/control) and investigator (investigator 1/investigator 2) and the random factor location nested under the diagnosis. The p-values were given for a two-sided problem. Figure 9 shows the original values of the percentage of anagen hairs for the di¡erent diagnoses and investigators with 95% con¢dence intervals of the means. The same is shown (Fig 10) for the length di¡erences between anagen and telogen hairs. The model of the percentage of anagen hairs explains 96% of the variance, the hair growth 85%. Both models are highly signi¢cant. Table II shows the model and the single e¡ects. For both variables the diagnosis is
70
Hair count
60
50
40
30
20 Cm
Eh
Kh
Th
Uh
Subject
Figure 5. Analysis of intraclass correlation. The intraclass correlation of one measurement in ¢ve volunteers (subjects) from two di¡erent investigators is shown for hair counts and for cumulative hair thickness (Fig 6).
6.0 Cumulative hair thickness (mm)
112
5.0
4.0
3.0
2.0 Cm
Eh
Kh
Th
Uh
Subject
Figure 6. Analysis of intraclass correlation. The intraclass correlation of one measurement in ¢ve volunteers (subjects) from two di¡erent investigators is shown for hair counts (Fig 5) and for cumulative hair thickness.
highly signi¢cant. These results show that an AGA-a¡ected scalp has a lower number of anagen hair follicles (Fig 9) and that these hair follicles grow slower (Fig 10), compared with hair follicles at the occiput. DISCUSSION
Numerous hair diseases such as the scarring alopecias, alopecia areata, or trichotillomania, usually do not need a quantitative method to evaluate the amount of hair shedding. Androgenetic e¥uvium, however, the most common form of hair loss, is typically di⁄cult to quantify and at present simple but reliable o⁄ce analytical procedures have not been developed. Although scalp biopsies can be justi¢ed in that microscopic examination of scalp skin a¡ected by AGA can identify and quantify any changes resulting from treatment, this invasive technique is often not suitable to monitor patients over a prolonged period of time. The classical trichogram is harmless and easy to use, but not reliable. AGA can be de¢ned as an androgen-dependent process in genetically predisposed individuals, where balding is due to continuous miniaturization of a¡ected hair follicles, changing large terminal hair follicles into small vellus-like hairs (Whiting et al, 1999; Ho¡mann and Happle, 2000a,b). Any successful treatment should therefore stop or reverse the process of hair follicle miniaturization, increase the number of terminal hair follicles, while reducing vellus hair counts (Whiting et al, 1999) or increase the number of anagen hairs (Van Neste et al, 2000). This concept is illustrated by the phase III studies of men with AGA treated with ¢nasteride (Kaufman et al, 1998). In these studies, macrophotographs were taken and hairs were counted. This technique produces counts of ‘‘visible’’ hairs, which means that tiny vellus-like hairs cannot be seen or counted. During treatment, however, these vellus-like hair follicles enlarge and subsequently increase the hair count results when the macrophotograph method is used. A major disadvantage of this technique is that it cannot monitor the expected continuous increase in hair thickness during treatment. As a consequence the phase III studies of men with AGA treated with ¢nasteride revealed that the increase in hair counts reaches a plateau after 1 y of treatment, whereas the hair coverage analyzed by global photographs increases continuously (Kaufman et al, 1998). This increase in hair coverage is due to an increase in hair thickness as shown by histologic examination (Whiting et al, 1999), the direct measurement of hair thickness (Steiner et al, 2000), and by a continuous increase in hair weight (Whiting, 2001). Although the Ludwig pattern of AGA in women di¡ers
VOL. 8, NO. 1 JUNE 2003
Table I. Trichoscan results from treated (¢nasteride) untreated men and controls 3 and 6 months after baseline
20
Variable
10
Group
Hair count
Control
0
AGA untreated AGA treated
-10 Diff. after 3 months
Diff. after 6 months
-20 11
11
5
control
Change in hair count
B
5
12
Control
12
AGA without
AGA with
treatment
treatment
AGA untreated AGA treated
140%
Time
df
t
p
After 3 mo After 6 mo After 3 mo After 6 mo After 3 mo After 6 mo After 3 mo
10 10 4 4 11 11 10
0.265 0.029 0.199 0.106 2.149 2.688 0.012
0.796 0.977 0.852 0.921 0.055 0.021 0.991
After 6 mo After 3 mo After 6 mo After 3 mo After 6 mo
10 4 4 11 11
0.540 0.739 1.161 2.414 3.433
0.601 0.501 0.310 0.034 0.006
120%
A
100%
80% Diff. after 3 months
60%
Diff. after 6 months N=
11
11
control
5
5
12
12
AGA without
AGA with
treatment
treatment
Figure 7. (a) Hair counts was analyzed for 6 mo in 11 volunteers without AGA, in ¢ve untreated men with AGA, and in 12 men treated with ¢nasteride (1 mg per day). In controls and untreated men we noticed no significant di¡erence in the number of hairs within the observation time of 6 mo (between the values at baseline and after 6 mo). In contrast those men treated with ¢nasteride showed a continuous increase (mean with 95% con¢dence interval) at 3 mo (p ¼ 0.055) and at 6 mo (p ¼ 0.021) in the number of hairs within the analyzed area compared with the values at baseline. (b) Hair counts was analyzed for 6 mo in 11 volunteers without AGA, in ¢ve untreated men with AGA, and in 12 men treated with ¢nasteride (1 mg per day). Quotients of hair count between measurements after 3 and 6 mo and ¢rst measurement for healthy subjects (controls), treated, and untreated patients with AGA (mean with 95% con¢dence interval).
in appearance (Ludwig, 1977) from the Hamilton pattern occurring in men, the pathophysiologic mechanisms seem to be the same, because female AGA patients treated with cyproterone acetate (Mortimer et al, 1984; Peereboom-Wynia et al, 1989; Dawber, 1995) or minoxidil (Price and Menefee, 1990; Rushton, 1993) experience an increase in hair thickness (Mortimer et al, 1984; Vanderveen et al, 1984; Peereboom-Wynia et al, 1989; Dawber, 1995) or hair weight (Price and Menefee, 1990). Therefore, a reliable hair counting method should primarily be able to calculate the number and thickness of hairs, which is stable within at least 1 cm above the scalp (Jackson et al, 1972; Hutchinson and Thompson, 1997), in a de¢ned area of the scalp. From a clinical perspective, hair thickness is very important (de Lacharriere et al, 2001), whereas the growth rate (mm per day) and the anagen/telogen ratio are of secondary importance. As early as 1964, Barman et al (1964) related a method that used optical contact microscopy to calculate these parameters, and much later Hayashi et al (1991) and recently D’Amico et al (2001) described a similar approach for the measurement of hair growth using optical microscopy and computer analysis; however, these investigators were unable to automate the process of calculation and measured the thickness of hairs visually with the cursor on the computer monitor. They calculated that results from di¡erent
Change in cumulative hair thickness (mm)
N=
Cumulative hair thickness
2.0
1.0
0.0
-1.0 Diff. after 3 months Diff. after 6 months
-2.0 N=
11
11
5
control
B Change in cumulative hair thickness
Change in hair count (95% CI)
A
113
TRICHOSCAN FOR HAIR MEASUREMENT
5
12
AGA without treatment
12
AGA with treatment
140%
120%
100%
80%
Diff. after 3 months
Diff. after 6 months
60% N = 11
11 control
5
5
AGA without treatment
12
12
AGA with treatment
Figure 8. (a) Cumulative hair thickness were analyzed for 6 mo in 11 volunteers without AGA, in ¢ve untreated men with AGA, and in 12 men treated with ¢nasteride (1 mg per day). Untreated men showed a continuous and signi¢cant decrease in the overall thickness of hairs 3 and 6 mo after the initial visit (baseline). In contrast those men treated with ¢nasteride showed in comparison with the baseline visit a continuous and signi¢cant increase in the hair thickness within the analyzed area after 3 mo (p ¼ 0.034) and 6 mo (p ¼ 0.006). (b) Hair counts and cumulative hair thickness was analyzed for 6 mo in 11 volunteers without AGA, in ¢ve untreated men with AGA, and in 12 men treated with ¢nasteride (1 mg per day). Quotients of cumulative hair thickness between measurements after 3 and 6 mo and ¢rst measurement for healthy subjects (controls), treated, and untreated patients with AGA (mean with 95% con¢dence interval).
investigators, but from the same image, di¡er by 7 8.4%, which makes such a semiautomatic method unsuitable for clinical practice. A similar approach has been tested using a phototrichogram,
114
HOFFMANN
JID SYMPOSIUM PRECEEDINGS
Table II. Trichoscan results for anagen hair and growth rate in AGA-a¡ected and una¡ected scalp
Anagen Hairs [%] (95% CI)
100
Anagen hairs 80
Hair growth 70
AGA
60 N=
df
F ratio
p
Model Diagnosis Investigator Image (diagnosis) Model Diagnosis Investigator Image (diagnosis)
34 1 1 31 34 1 1 31
17.90 57.60 0.15 7.18 4.95 15.10 2.00 3.40
o0.0001 o0.0001 0.702 o0.0001 o0.0001 o0.001 0.168 o0.001
control 18
14
18
A
13
B
Investigator
Figure 9. Analysis of anagen/telogen ratio. While de¢ning anagen hairs as growing and telogen hairs as nongrowing hairs 3 d after shaving, the TrichoScan is able to calculate a digital trichogram. This ¢gure illustrates the results (mean with 95% con¢dence interval) from two investigators, who analyzed in 18 volunteers with AGA the proportion of anagen hairs at the vertex and at the occiput (only 14 images). Compared with the occiput, the AGA-a¡ected scalp reveals a decreased number of anagen hairs. Both investigators produced similar results.
.42
Hair Growth [mm/day] (95% CI)
E¡ect 90
.40 .38 .36 .34 AGA
.32
control
.30 N=
18
14
18
A
13
B
Investigator
Figure 10. Analysis of hair growth rate. This ¢gure illustrates the results (mean with 95% con¢dence interval) from two investigators, who analyzed in 18 volunteers with AGA the and hair growth rate at the vertex and at the occiput (only 14 images). Compared with the occiput, the AGAa¡ected scalp reveals slower growing anagen hairs. Both investigators produced similar results.
which has proven to be a suitable, noninvasive tool to monitor hair growth phases in situ. This technique has been improved with image analysis (Van Neste et al, 1989) and later using immersion oil and digital contrast enhancement (Van Neste et al, 1992). Despite markedly improved images and more accurate quantitative data (Van Neste et al, 1992), however, automated analysis was not possible. Until now, analysis of images has been tedious and time consuming. Unsuccessful attempts to automate the process have been reported several times (Pel¢ni and Calligaro, 1986; Hayashi et al, 1991; Van Neste et al, 1992). The major problem has been that scalp hair follicles grow in groups (follicular units) rather than singly, and therefore neighboring hair follicles typically overlap or may be aligned in parallel. Furthermore, photographic analysis software requires good contrast between the hair follicles and the scalp skin, and the fact that many hairs lose their natural pigmentation due to aging or AGA, makes them much more di⁄cult to
detect. We have overcome this di⁄culty by coloring the hairs prior to taking the images, without any negative e¡ect on the collected data. Furthermore, we have created automatic software for the analysis of the aforementioned parameters of hair growth. Because this described technique is a modi¢ed and computerized trichogram we called it TrichoScan. Images were taken with a video system using ELM. ELM is a standard procedure used in analyzing melanocytic nevi (Steiner et al, 1987; Braun-Falco et al, 1990; Kreusch et al, 1992; Krischer et al, 1999), and many dermatologists in Europe already use ELM systems in daily clinical practice. These devices produce high-quality and reproducible digital images, because the images are always taken at the same distance from the skin surface. Our results suggest that ELM systems can be used to evaluate patients complaining of androgenetic e¥uvium and for monitoring responses to therapy. Variations that normally occur in hair length, weight, thickness, etc. can be assayed either with respect to standardized values, or by comparing measurements made on two or more occasions over a given period of time. Similar means of assay must be employed to study changes in hair growth that may occur with regard to age or illness. The margin of error of the techniques and the instruments employed should be smaller than the magnitude of the variations to be measured. As our results show, the TrichoScan ful¢lls these criteria and has advantages over standard procedures used thus far for hair measurements. 1 It is investigator independent. In other studies using the unit area trichogram, substantial di¡erences between investigators were noted. In these studies a signi¢cantly larger mean total hair counts were reported from experienced compared with inexperienced observers (Rushton et al, 1989). Our results show that this is not the case for the TrichoScan technique. 2 Many methods are not strictly validated. The hair weight test is a good example where the hair is clipped in a de¢ned target area; however, the sample error for di¡erent investigators is unknown. This is mainly due to the methodology itself, because once hairs are clipped a second investigator cannot clip the same area again to assess the reproducibility of the method. In contrast, the TrichoScan is highly validated with de¢ned values for intraclass correlation between the same and di¡erent investigators. This is of crucial importance for clinical studies. In cases where the expected di¡erences between placebo and actively treated patients is known, the minimum number of patients necessary to prove this di¡erence can be calculated. 3 Some methods are associated with considerable discomfort to the patient such as the repeated plucking of hairs required by the trichogram technique. The TrichoScan relies on a small analyzed area of the scalp, which is afterwards barely visible. The tiny tattoo is the only discomfort patients will recognize. 4 Some methods to count hairs are tedious and time-consuming. By contrast, TrichoScan can be used by an experienced operator within 8^12 min ‘‘hands on’’. 5 The amount of equipment is small. Many dermatologists already have ELM systems and these physicians would only need the TrichoScan software.
VOL. 8, NO. 1 JUNE 2003
TRICHOSCAN FOR HAIR MEASUREMENT
6 Many methods only count pigmented hair shafts.Via staining the hairs prior the analysis, the TrichoScan is able to analyze fair and gray hairs as well. 7 Most techniques require a rather long observation time to be able to detect signi¢cant di¡erences between groups of patients. In contrast, the TrichoScan allows the detection of a statistically signi¢cant increase in hair thickness even after 3 mo of treatment in a rather small cohort of patients. Therefore, the overall advantage of the TrichoScan is that this technique can be used for clinical studies to compare placebo vs treatment or to compare di¡erent capacities of di¡erent hair growth promoting substances. This technique can be used to study AGA or other forms of di¡use hair loss, Moreover, it can be adopted to study the e¡ect of drugs and laser treatment on hypertrichosis and hirsutism. The work of U. Ellwanger and H. Lˇdtke (Datenanalyze und Angewandte Informatik GbR [www.datinf.de], Brunnenstr. 14, 72074 Tˇbingen, Germany) in programming the software is greatfully appreciated.
REFERENCES Barman JM, Pecoraro V, Astore I: Method, technic and computations in the study of the trophic state of human scalp hair. J Invest Dermatol 42:421^425, 1964 Barth JH, Rushton DH: Measurement of hair growth. In: Serudp J, Jemee GBE (eds). Non-Invasive Methods and the Skin, Vol. 1. Ann Arbor: CRP Press, 1995: pp 543^548 Blume-Peytavi U, Orfanos CE: Microscopy of the hair. In: Serudp J, Jemee GBE (eds). Non-Invasive Methods and the Skin, Vol. 1. Ann Arbor: CRP Press, 1995: pp 549^554 Bouhanna P: The phototrichogram, a macrophotographic study of the scalp. Bioengineer Skin 3:265, 1985 Braun-Falco O, Stolz W, Bilek P, Merkle T, Landthaler M: The dermatoscope: Hautarzt. 41:131^136, 1990 Can¢eld D: Photographic documentation of hair growth in androgenetic alopecia. Dermatol Clin 14:713^721, 1996 D’Amico D, Vaccaro M, Guarneri F, Borgia F, Cannavo S, Guarneri B: Phototrichogram using videomicroscopy: A useful technique in the evaluation of scalp hair. Eur J Dermatol 11:17^20, 2001 Dawber RPR: Haarwachstumsdiagnostik und das Haarwachstum stimulierende Substanzen. In: Tebbe B, Goerdt S, Orfanos CE (eds). Dermatologie, Heutiger Stand. Stuttgart: Georg Thieme Verlag, 1995:pp 243^248 Friedel J, Will F, Grosshans E: Phototrichogram. Adaptation, standardization and applications. Ann Dermatol Venereol 116:629^636, 1989 Guarrera M, Ciulla MP: A quantitative evaluation of hair loss: The phototrichogram. J Appl Cosmetol 4:61^66, 1986 Hamilton JB: Patterned loss of hair in man: Types and incidence. Ann N Y Acad Sci 53:708^728, 1951 Hayashi S, Miyamoto I, Takeda K: Measurement of human hair growth by optical microscopy and image analysis. Br J Dermatol 125:123^129, 1991 Headington JT: Transverse microscopic anatomy of the human scalp. A basis for a morphometric approach to disorders of the hair follicle. Arch Dermatol 120:449^456, 1984 Ho¡mann R, Happle R: Current understanding of androgenetic alopecia. Part I: Etiopathogenesis. Eur J Dermatol 10:319^327, 2000a Ho¡mann R, Happle R: Current understanding of androgenetic alopecia. Part II: Clinical aspects and treatment. Eur J Dermatol 10:410^417, 2000b Hutchinson PE,Thompson JR: The cross-sectional size and shape of human terminal scalp hair. Br J Dermatol 136:159^165, 1997
115
Jackson D, Church RE, Ebling FJ: Hair diameter in female baldness. Br J Dermatol 87:361^367, 1972 Kaufman KD, Olsen EA, Whiting D, et al: Finasteride in the treatment of men with androgenetic alopecia. Finasteride Male Pattern Hair Loss Study Group. J Am Acad Dermatol 39:578^589, 1998 Kreusch J, Rassner G, Trahn C, Pietsch-Breitfeld B, Henke D, Selbmann HK: Epiluminescent microscopy: A score of morphological features to identify malignant melanoma. Pigment Cell Res Suppl 2:295^298, 1992 Krischer J, Braun RP, Toutous-Trellu L, Saurat JH, Pechere M: Kaposi’s sarcoma: A new approach of lesional follow-up using epiluminescent light microscopy. Dermatology 198:420^422, 1999 de Lacharriere O, Deloche C, Misciali C, et al: Hair diameter diversity: A clinical sign re£ecting the follicle miniaturization. Arch Dermatol 137:641^646, 2001 Ludwig E: Classi¢cation of the types of androgenetic alopecia (common baldness) occurring in the female sex. Br J Dermatol 97:247^254, 1977 Maguire HC, Kligman AM: Hair plucking as a diagnostic tool. J Invest Dermatol 43:77^79, 1964 Mortimer CH, Rushton H, James KC: E¡ective medical treatment of common baldness in women. Clin Exp Dermatol 9:342^350, 1984 Pecoraro V, Astore I, Barman JM: Growth rate and hair density of the human axilla. A. Comparative study of normal males and females and pregnant and postpartum females. J Invest Dermatol 56:362^365, 1971 Peereboom-Wynia JD, van der Willigen AH, van Joost T, Stolz E: The e¡ect of cyproterone acetate on hair roots and hair shaft diameter in androgenetic alopecia in females. Acta Derm Venereol 69:395^398, 1989 Pel¢ni C, Calligaro A: Evaluation of hair growth by means of morphometric computerized analysis. J Appl Cosmetol 4:67^76, 1986 Price VH, Menefee E: Quantitative estimation of hair growth. I. Androgenetic alopecia in women: E¡ect of minoxidil. J Invest Dermatol 95:683^687, 1990 Rushton DH: Management of hair loss in women. Dermatol Clin 11:47^53, 1993 Rushton H, James KC, Mortimer CH: The unit area trichogram in the assessment of androgen-dependent alopecia. Br J Dermatol 109:429^437, 1983 Rushton DH, Unger WP, Cotterill PC, Kingsley P, James KC: Quantitative assessment of 2% topical minoxidil in the treatment of male pattern baldness. Clin Exp Dermatol 14:40^46, 1989 Rushton DH, de Brouwer B, de Coster W, van Neste DJ: Comparative evaluation of scalp hair by phototrichogram and unit area trichogram analysis within the same subjects. Acta Derm Venereol 73:150^153, 1993 Saitoh M, Uzuka M, Sakamoto M: Human hair cycle. J Invest Dermatol 54:65^81, 1970 Steiner A, Pehamberger H, Wol¡ K: Improvement of the diagnostic accuracy in pigmented skin lesions by epiluminescent light microscopy. Anticancer Res 7:433^ 434, 1987 Steiner D, Bedin V, Pasello RR: Hair shaft diameter evaluation in AGA before and after ¢nasteride 1 mg per day. Poster Nr 284. Annual Meeting of the AAD 2000 Van Neste D: Human scalp hair growth and loss evaluation methods: Is there simple and reliable method? Exp Dermatol 8:299^301, 1999 Van Neste DJ: Contrast enhanced phototrichogram (CE-PTG): An improved noninvasive technique for measurement of scalp hair dynamics in androgenetic alopeciavalidation study with histology after transverse sectioning of scalp biopsies. Eur J Dermatol 11:326^331, 2001 Van Neste DJJ, Dumrotier M, De Coster W: Phototrichogram analysis. Technical aspects and problems in relation with automated quantitative evaluation of hair growth by computer-assisted image analysis. In: Van Neste DJJ, Lachapelle JM, Antoine JL (eds). Trends in Human Hair Growth and Alopecia Research. Dordrecht: Kluwer, 1989:pp 155^165 Van Neste DJJ, Dumrtier M, de Brouwer B, De Coster W: Scalp immersion proxigraphy (SIP): An improved imaging technique for phototrichogram analysis. J Eur Acad Dermatol Venereol 1:187^191, 1992 Van Neste DJ, de Brouwer B, De Coster W: The phototrichogram: Analysis of some technical factors of variation. Skin Pharmacol 7:67^72, 1994 Van Neste D, Fuh V, Sanchez-Pedreno P, et al: Finasteride increases anagen hair in men with androgenetic alopecia. Br J Dermatol 143:804^810, 2000 Vanderveen EE, Ellis CN, Kang S, et al: Topical minoxidil for hair regrowth. J Am Acad Dermatol 11:416^421, 1984 Whiting DA: Advances in the treatment of male androgenetic alopecia: A brief review of ¢nasteride studies. Eur J Dermatol 11:332^334, 2001 Whiting DA, Waldstreicher J, Sanchez M, Kaufman KD: Measuring reversal of hair miniaturization in androgenetic alopecia by follicular counts in horizontal sections of serial scalp biopsies: Results of ¢nasteride 1 mg treatment of men and postmenopausal women. J Invest Dermatol Symp Proc 4:282^284, 1999