This website is intended exclusively for healthcare professionals and medical device manufacturers.
By accessing this site, you confirm that you belong to one of its of these categories and understand that the information provided is specifically intented for a professional audience.
"The expectation for any AI tool collecting data for decision making is that it must be clinically validated using diverse dataset, and that it must comply with Health Insurance Portability and Accountability Act (HIPAA)"
With background in chemical engineering, Fauad comes with more than 24 years of experience in various leadership roles in the Biopharma industry. Fauad became known for developing the first botulinum neurotoxin type E program at Bonti, where he was the co-founder and CEO, before its acquisition by Allergan. His vision and leadership, along with deploying a capital efficient and lean model, was key to Bonti’s fast progress in a very short time. Prior to that, he served in various leadership roles at Allergan, including Global Product Development teams of top tier neurotoxin (including Botox®) and ocular programs, establishing overall drug development strategy from discovery to commercialization. He also has patents related to drug development in various fields and has authored research articles in peer reviewed academic journals and chapters for scientific books.
Fauad is currently serving as an executive, board member, advisor, and investor with the focus on therapeutics, digital health/AI, and medical devices. He is also a consultant in the medical aesthetic field, advising global clients on both business and development strategies. Fauad is a board member of a business accelerator, Octane, and serves on the organizing committee of Aesthetic Tech Forum and Medical Aesthetics Injectable Summit.
Growing up in Silicon Valley, Fauad learned leadership skills at an early age through competitive sports. He won several awards while playing squash/cricket at regional and national level and represented USA National Cricket team in international competitions. Fauad received his graduate degrees from the University of Maryland.
Digital tools have long been part of skincare, especially in medical dermatology. Today, we have FDA-approved, AI-powered tools capable of classifying skin lesions and even detecting melanoma. On the consumer side, several beauty apps—often created by product companies—offer cosmetic recommendations and provide somewhat subjective skin condition analyses. Although the term “AI” is often used loosely in these applications, there is no denying its potential in skincare, particularly regarding visual perception. One critical concern, however, is that these algorithms can harbor biases due to unbalanced training datasets that don’t fully represent all populations.
In medical aesthetics, there remains a strong need for a comprehensive digital platform that benefits all key stakeholders: practitioners, patients, and product developers. The first step is achieving an accurate and consistent way to assess skin conditions and their severity. Numerous studies highlight significant variability in how different physicians diagnose dermatological conditions. This challenge is compounded in facial aesthetics research—such as when developing neurotoxins to treat wrinkles—because paper-based photo-numeric scales (often four-point scales ranging from “severe” to “none”) are still widely used. These proprietary scales guide crucial decisions about a product’s performance, yet they often introduce inconsistent scoring among practitioners. When patient self-assessments are factored in (a regulatory requirement), variability grows even further, frequently leading to inconsistent outcomes and frustration for product developers. This discrepancy can force costly study repeats and waste valuable time.
Artificial intelligence can help remove subjectivity by standardizing the process of scoring wrinkles. Although paper-based photo-numeric scales are validated through intensive image annotations by seasoned aesthetics physicians, that same annotated data can also train AI algorithms. A Deep Convolutional Neural Network (DCNN) can classify images into established categories such as wrinkle severity, and it can easily be extended to include the six-point Fitzpatrick scale, age group, gender, ethnicity, and more. This objectivity and standardization enhance both accuracy and consistency for practitioners, allowing AI algorithms to match or exceed human-level accuracy in assessing skin conditions.
As more data is collected over time, the machine learning components of the AI can evolve to include powerful back-end analytics. Researchers and developers can categorize data to track which products were used, how many units were administered, treatment frequency, onset of effects, and duration. This information paves the way for creating highly personalized treatment plans—especially as today’s average patient profile spans younger individuals, non-Caucasian backgrounds, and males, among others. Currently, many practitioners still rely on anecdotal experience to guide treatments, which can yield uneven results. With a predictive model informed by a large, diverse dataset, new and experienced practitioners alike can benefit from treatment suggestions that are tailored to individual attributes. Predictive modeling can also illustrate how a patient’s skin might change over time, with or without treatment, and help maintain desired outcomes by recommending an ideal treatment paradigm.
Patients, too, benefit from tracking their progress through standardized “selfies” and periodic assessments. Just as fitness apps help users stay on track through clear metrics and real-time feedback, democratizing severity scales via a digital platform empowers individuals to manage and control their skin outcomes. With quick, private access to their personalized “score,” patients can monitor improvements, see when it’s time for another treatment, and stay motivated as they witness their progress firsthand. They can also report concerns or successes through the platform, giving practitioners prompt feedback that enhances both satisfaction and retention.
On the clinical side, advanced imaging tools (2D/3D, audio, and visual) can create immersive experiences for researchers, investigators, and patients—offering real-time insights into product safety and efficacy, including onset of action and duration of results. Allergan Aesthetics has developed a proprietary clinical research platform for their products called Aptios, where study subjects can submit images directly to their clinical study teams. There are similar platforms available that other companies who may not have similar resources can use to achieve similar outcomes much more efficiently. On the commercial side, collecting real-world data following a product’s launch is essential for understanding safety and long-term effectiveness, refining marketing strategies, and supporting potential future claims. By feeding both clinical and commercial data into deep machine learning models, companies can develop ever more effective and personalized products that ultimately deliver better outcomes for patients.
While there are no regulatory specific guidelines tailored to the use of AI in medical aesthetics, earlier this year USA FDA released a draft guidance for AI enabled devices. The expectation for any AI tool collecting data for decision making is that it must be clinically validated using diverse dataset, and that it must comply with Health Insurance Portability and Accountability Act (HIPAA). In Europe Medical Device Regulation (MDR) and the AI Act provide similar oversight, and emphasize demonstrating data integrity, protection, secure handling, bias elimination, and mechanism for human oversight.
Yoelin (2022). The Use of a Novel Artificial Intelligence Platform for the Evaluation of Rhytids. Aesthet Surg J. 2022 Oct 13;42(11):NP688-NP694. doi: 10.1093/asj/sjac200.
