Development and qualification of a machine learning algorithm for automated hair counting.

OBJECTIVE: Determining the amount of hair on the scalp has always been an important metric of patient satisfaction for hair growth and hair retention technologies. While simple in concept, this measurement is a difficult, resource intensive task for the dermatologist and the research scientist. Specifically, counting and measuring hair in phototrichogram images is very time consuming and labour intensive. Due to cost, often only a fraction of available images is manually analysed. There is a need for an automated method that can significantly increase speed and throughput while reducing the cost of counting and measuring hair in phototrichogram images. METHODS: Recent advances in machine learning and deep convolutional neural networks (deep learning) have led to a revolution in the analysis of image, video, speech, text and other sensor data. Image diagnostics have seen remarkable improvements with completely automated methods outperforming both human experts and human-engineered analysis methods. Deep learning methods can also provide speed and cost benefits. To enable use of a deep learning, we created a data set of 288 manually annotated phototrichogram images with marked location and length of each hair (the training dataset). We designed a custom neural network architecture and custom image processing algorithms to best utilize the available training data and to maximize performance for hair counting and length measurement. The performance of the algorithm was qualified by comparing hair count and length measurements to an independent ground truth method, the semi-manual Canfield's Hair Metrix method. RESULTS: Leveraging deep neural networks, we have developed capability to apply machine learning to reduce the time needed to acquire data from phototrichograms of patients' scalp from months to seconds. Our algorithm enables fast and fully automated hair counting and length measurement. The algorithm shows high agreement with human manually assisted analysis (ground truth). CONCLUSIONS: We have trained and deployed an algorithm utilizing this technology and have demonstrated the reproducibility, accuracy and speed of this algorithm that, once deployed, requires little to no recurring cost or manual intervention for its operation. The method allows fast analysis of large number of images, reducing study cost and significantly reducing study analysis time.

Zinc pyrithione inhibits yeast growth through copper influx and inactivation of iron-sulfur proteins.

Zinc pyrithione (ZPT) is an antimicrobial material with widespread use in antidandruff shampoos and antifouling paints. Despite decades of commercial use, there is little understanding of its antimicrobial mechanism of action. We used a combination of genome-wide approaches (yeast deletion mutants and microarrays) and traditional methods (gene constructs and atomic emission) to characterize the activity of ZPT against a model yeast, Saccharomyces cerevisiae. ZPT acts through an increase in cellular copper levels that leads to loss of activity of iron-sulfur cluster-containing proteins. ZPT was also found to mediate growth inhibition through an increase in copper in the scalp fungus Malassezia globosa. A model is presented in which pyrithione acts as a copper ionophore, enabling copper to enter cells and distribute across intracellular membranes. This is the first report of a metal-ligand complex that inhibits fungal growth by increasing the cellular level of a different metal.

Isolation and expression of a Malassezia globosa lipase gene, LIP1.

Dandruff and seborrheic dermatitis (D/SD) are common hyperproliferative scalp disorders with a similar etiology. Both result, in part, from metabolic activity of Malassezia globosa and Malassezia restricta, commensal basidiomycete yeasts commonly found on human scalps. Current hypotheses about the mechanism of D/SD include Malassezia-induced fatty acid metabolism, particularly lipase-mediated breakdown of sebaceous lipids and release of irritating free fatty acids. We report that lipase activity was detected in four species of Malassezia, including M. globosa. We isolated lipase activity by washing M. globosa cells. The isolated lipase was active against diolein, but not triolein. In contrast, intact cells showed lipase activity against both substrates, suggesting the presence of at least another lipase. The diglyceride-hydrolyzing lipase was purified from the extract, and much of its sequence was determined by peptide sequencing. The corresponding lipase gene (LIP1) was cloned and sequenced. Confirmation that LIP1 encoded a functional lipase was obtained using a covalent lipase inhibitor. LIP1 was differentially expressed in vitro. Expression was detected on three out of five human scalps, as indicated by reverse transcription-PCR. This is the first step in a molecular description of lipid metabolism on the scalp, ultimately leading toward a test of its role in D/SD etiology.