Qualification of a new and precise automatic tool for the assessment of hair diameters in phototrichograms

Machine learning (ML) techniques as effective methods for evaluating hair and skin assessments: A systematic review

Machine Learning (ML) techniques provide the ability to effectively evaluate and analyze human skin and hair assessments. The aim of this study is to systematically review the effectiveness of applying Machine Learning (ML) methods and Artificial Intelligence (AI) techniques in order to evaluate hair and skin assessments. PubMed, Web of Science, IEEE Xplore, and Science Direct were searched in order to retrieve research publications between 1 January 2010 and 31 March 2020 using appropriate keywords such as "hair and skin analysis." Following accurate screening, 20 peer-reviewed publications were selected for inclusion in this systematic review. The analysis demonstrated that prevalent Machine Learning (ML) methods comprised of Support Vector Machine (SVM), k-nearest Neighbor, and Artificial Neural Networks (ANN). ANN's were observed to yield the highest accuracy of 95% followed by SVM generating 90%. These techniques were most commonly applied for drafting framework assessments such as that of Melanoma. Values of parameters such as Sensitivity, Specificity, and Area under the Curve (AUC) were extracted from the studies and with the help of comparisons, relevant inferences were also made. ANN's were observed to yield the highest sensitivity of 82.30% as well as a 96.90% specificity. Hence, with this systematic review, a summarization of the studies was drafted that encapsulated how Machine Learning (ML) techniques have been employed for the analysis and evaluation of hair and skin assessments.

Invention of automated numerical algorithm adopting binarization for the evaluation of scalp hair coverage: An image analysis providing a substitute for phototrichogram and global photography assessment for hair diseases

BACKGROUND

The efficacy of therapeutic modalities for hair disease can be evaluated globally by photo assessment and more precisely by phototrichogram (PTG). However, the latter procedure is laborious, time consuming, subject to inter-observer variation, and requires hair clipping.

OBJECTIVE

To establish an automated and patient/investigator friendly methodology enabling quantitative hair amount evaluation for daily clinical practice.

METHODS

A novel automated numerical algorithm (aNA) adopting digital image binarization (i.e., black and white color conversion) was invented to evaluate hair coverage and measure PTG parameters in scalp images. Step-by-step improvement of aNA was attempted through comparative analyses of the data obtained respectively by the novel approach and conventional PTG/global photography assessment (GPA).

RESULTS

For measuring scalp hair coverage, the initial version of aNA generally agreed with the cumulative hair diameter as assessed using PTG, showing a coefficient of 0.60. However, these outcomes were influenced by the angle of hair near the parting line. By integrating an angle compensation formula, the standard deviation of aNA data decreased from 5.7% to 1.2%. Consequently, the coefficient of determination for hair coverage calculated using the modified aNA and cumulative hair diameter assessed by PTG increased to 0.90. Furthermore, the change in hair coverage as determined by the modified aNA protocol correlated well with changes in the GPA score of images obtained using clinical trials.

CONCLUSION

The novel aNA method provides a valuable tool for enabling simple and accurate evaluation of hair growth and volume for clinical trials and for treatment of hair disease.

Evaluation of Hair Density and Hair Diameter in the Adult Thai Population Using Quantitative Trichoscopic Analysis

The data of hair density and hair diameter in the Asian population, especially in Thais, are limited. We aimed to evaluate hair density and hair diameter of members of the Thai population at different scalp sites and to determine the effect of sex and aging as well as to compare the results with those in groups of other ethnicities. Healthy Thais whose hair examination findings were normal were evaluated. Two hundred and thirty-nine subjects participated in this study, of whom 79 were male and 160 were female. Hair density and hair diameter were analyzed at four different scalp sites using quantitative trichoscopic analysis. The highest hair density in Thais was observed in the vertex area. Hair densities at four different scalp sites were significantly different from one another; only hair density at the vertex site showed no significant difference from that in the occipital area. In contrast, hair diameter did not show any statistically significant differences for the different sites. We observed decreased mean hair density with increasing age and found statistically significant differences between participants in their 20s and those in their 60s, while hair diameter remained consistent. Comparing our results with a previous study in other ethnicities, the hair densities in Asians are generally lower. In conclusion, hair density in the Thai population varies at different scalp sites. Aging is a factor in declining hair density. Asians have a lower hair density compared to Caucasian and African populations.

Significant relationship between temporal hair loss and other scalp areas in female pattern hair loss

Female pattern hair loss affects the central scalp, sparing the frontal hairline. The temporal area can also be affected by hair loss. We investigated the degree of temporal hair loss and correlation of other sites of scalp hair loss in Korean female pattern hair loss patients. A total of 109 women with female pattern hair loss were enrolled in this retrospective analysis. We measured hair density and thickness in five scalp sites including the frontal, vertex, occipital and bilateral temporal areas by phototrichogram. Frontal and vertex area hair loss were classified according to the Basic and Specific (BASP) classification, and temporal scalp and occiput areas were also assessed. Eighty-nine patients showed temporal hair loss. The mean of the hair density was lowest in the temporal area among all scalp areas. Total and thick hair densities of the frontal scalp were correlated with those of the vertex, temporal scalp and occiput in descending order, and hair thickness of the frontal scalp was more related with that of the temporal scalp than the vertex. In this study, temporal involvement is evident in female pattern hair loss. We suggest that temporal involvement should be added to pattern hair loss classification, especially BASP classification.

Characteristic findings by phototrichogram in southern Chinese women with Female pattern hair loss

OBJECTIVES

To investigate the characteristics of hairs in Female pattern hair loss (FPHL) patients and healthy females in Southern China.

MATERIALS AND METHODS

Three fundamental hair parameters in different scalp areas of 90 Southern Chinese FPHL patients and 83 healthy controls were analyzed by phototrichogram.

RESULTS

Female pattern hair loss patients showed reduced hair density, hair diameter, and terminal/vellus hair ratio. The reduction correlated with the severity of Ludwig staging. Midscalp was the most affected area in FPHL, but occipital and temporal sites were also involved. In normal women, the highest hair density was observed in midscalp, followed by occipital and temporal areas. Peak hair density at midscalp sites was reached at 20s group, then declined with age. Maximum hair diameter at midscalp and occipital sites occurred in 40s group. Terminal/vellus hair ratio tended to increase with age and peak on 50-60s group.

CONCLUSION

Reduced hair density and hair diameter, and miniaturization of hair follicles are the characteristics of FPHL in Southern Chinese women. Occipital and temporal sites are also affected in FPHL. Age-associated changes might have an influence on the hair condition. The values of hair parameters obtained in this study will help to establish reference data for better evaluation of hair disorders.

Towards a body hair atlas of women of caucasian ethnicity

OBJECTIVE

A preliminary study was conducted in 17 female volunteers (mean age 29.8 years) to gain deeper insights into the characteristics of terminal Caucasian female body hair of different body parts. The focus on Caucasian women was driven by the high number of different scalp hair phenotypes in this ethnicity and intended to identify relevant differences between body areas to improve body hair removal approaches.

METHODS

Multiple growth parameters and structural parameters were assessed for hair on the upper arm, forearm, upper leg, lower leg, axilla and intimate area and compared to scalp data.

RESULTS

In particular, macroscopic and much less microscopic or hair surface properties differ strikingly in the investigated body areas. Hair density on the body is much lower than on scalp with the highest hair density in the axilla and intimate area. Multihair follicular units are described for scalp but were also found to a smaller proportion in the axilla and the intimate area. Substantial percentages of hair triplets are only found on the scalp and intimate area. Hair diameter is highest in the intimate area, followed by axillary and lower leg hair and correlates with a faster hair growth rate. The angle of emerging hair is smallest in the intimate area, axilla and on the lower leg. Hair shafts on the lower leg and in the axilla have most overlapping cuticle layers, but independent of body region, no significant differences in the mean thickness of cuticle layers were detectable. In addition, no differences were found in the mean distance between cuticle layer edges along the hair shaft and the hair surface roughness. Hair on the scalp, forearm, upper arm and upper leg had an almost round shape, whereas hair of the lower leg, intimate area and axilla had more elliptical shape. Hairs on the arm showed the highest luminance values and no visible medulla. The darkest hairs were in the axilla and intimate area containing the highest level of visible medulla in hair shafts.

CONCLUSION

To our knowledge, this is the first systematic study comparing terminal hair properties in all cosmetically relevant body regions in Caucasian women.

Cross-section Trichometry: A Clinical Tool for Assessing the Progression and Treatment Response of Alopecia

Background:

To properly assess the progression and treatment response of alopecia, one must measure the changes in hair mass, which is influenced by both the density and diameter of hair. Unfortunately, a convenient device for hair mass evaluation had not been available to dermatologists until the recent introduction of the cross-section trichometer, which directly measures the cross-sectional area of an isolated bundle of hair.

Objective:

We sought to evaluate the accuracy and sensitivity of the HairCheck^® device, a commercial product derived from the original cross-section trichometer.

Materials and Methods:

Bundles of surgical silk and human hair were used to evaluate the ability of the HairCheck^® device to detect and measure small changes in the number and diameter of strands, and bundle weight.

Results:

Strong correlations were observed between the bundle's cross-sectional area, displayed as the numeric Hair Mass Index (HMI), the number of strands, the silk/hair diameter, and the bundle dry weight.

Conclusion:

HMI strongly correlated with the number and diameter of silk/hair, and the weight of the bundle, suggesting that it can serve as a valid indicator of hair mass. We have given the name cross-section trichometry (CST) to the methodology of obtaining the HMI using the HairCheck^® system. CST is a simple modality for the quantification of hair mass, and may be used as a convenient and useful tool to clinically assess changes in hair mass caused by thinning, shedding, breakage, or growth in males and females with progressive alopecia or those receiving alopecia treatment.

Toponomics method for the automated quantification of membrane protein translocation

Background

Intra-cellular and inter-cellular protein translocation can be observed by microscopic imaging of tissue sections prepared immunohistochemically. A manual densitometric analysis is time-consuming, subjective and error-prone. An automated quantification is faster, more reproducible, and should yield results comparable to manual evaluation. The automated method presented here was developed on rat liver tissue sections to study the translocation of bile salt transport proteins in hepatocytes. For validation, the cholestatic liver state was compared to the normal biological state.

Results

An automated quantification method was developed to analyze the translocation of membrane proteins and evaluated in comparison to an established manual method. Firstly, regions of interest (membrane fragments) are identified in confocal microscopy images. Further, densitometric intensity profiles are extracted orthogonally to membrane fragments, following the direction from the plasma membrane to cytoplasm. Finally, several different quantitative descriptors were derived from the densitometric profiles and were compared regarding their statistical significance with respect to the transport protein distribution. Stable performance, robustness and reproducibility were tested using several independent experimental datasets. A fully automated workflow for the information extraction and statistical evaluation has been developed and produces robust results.

Conclusions

New descriptors for the intensity distribution profiles were found to be more discriminative, i.e. more significant, than those used in previous research publications for the translocation quantification. The slow manual calculation can be substituted by the fast and unbiased automated method.