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

Abstract

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.

Scalp application of antioxidants improves scalp condition and reduces hair shedding in a 24‐week randomized, double‐blind, placebo‐controlled clinical trial

“…A trained operator used the imaging system. Total hair counts were analysed by a blinded operator at Canfield Scientific and separately with a novel machine-learning algorithm for counting hairs [25]. This method was qualified with data from previous hair quantification studies.…”

Comparison Between Trichoscopic and Histopathological Evaluations of Hair Parameters

“…With the development of artificial intelligence, the machine learning algorithm may reduce manual assistance and improve accuracy. 18 …”

Scalp application of the antioxidant piroctone olamine reduces hair shedding in an 8‐week randomized, double‐blind, placebo‐controlled clinical study

“…For the subjects on the leave-on treatment, phototrichogram assessment was performed. All phototrichogram analysis was performed using the automated AI algorithm previously described [2,10]. The details of the measurements follow.…”

“…A trained operator used the imaging system. Hair counts were analysed with a novel machinelearning algorithm for counting hairs [10]. This method was qualified with data from previous hair quantification studies.…”

“…Phototrichogram images within the scalp hair-clipped areas were analysed by an in-house automated algorithm [1,10]. Total hair counts were greater in subjects using the piroctone olamine (PO) shampoo versus those using the placebo shampoo (Figure 3a).…”