Automated Detection of Street Tobacco Displays - Isha Chaturvedi's Prodigious Work in the Public Sector Inbox

 Isha Chaturvedi

With the various rapid developments pertaining to a multitude of lifestyle and technological shifts propelling dramatic changes in our lives, it is an urgent call for the global health effort to transcend national boundaries and work collaboratively. Interesting innovations and researches are cultivating on all continents - all committed and streamlined in the direction of making human lives better. One of such marvelous works in the territory of public health has been accomplished by Isha Chaturvedi, an NYU graduate in Data Science, in a team of five members.

About the Work

The team's prodigious efforts targeted tracking down the vendors that have tobacco advertising displays in an automated way (employing data science techniques) to observe their vicinity from schools, areas of communities of color, and low-income communities. Upon the fulfillment of an extensive literature study, the team concluded that the lack of empirical evidence linking product exposure to behavior remains a key obstacle in the adoption of additional restrictions on point-of-sale tobacco advertising.
Since the advent of modern advertising in the 1920s, as the team noted, tobacco marketers have evolved and worked ceaselessly to thrust young audiences into smoking early. The consequences of these companies' relentless and persuasive messaging through mainstream media channels like movies and magazines became increasingly apparent over the years.

Hence, the purpose of undertaking this project was to map point-of-sale tobacco marketing practices across New York City using automated detection of tobacco signage in street-level imaging data. They ultimately put forward a proof-of-concept for measuring the exposure of at-risk communities to tobacco displays. The team also encompassed a study to comprehend the health behavioral aspects, attitudes about the environment as they occur within the residents' day-to-day lives, along with the outcomes generated.

She worked on this project during her final year at New York University. Amalgamating her efforts with her four other fellows, this project was sponsored by Dr. Tom Kirchner of NYU mHealth Lab, a research group within the NYU College of Global Public Health. This lab functions to leverage the power of cellphones to collect data about neighborhoods and experiences, including exposure to different risk and protective factors such as tobacco point-of-sale.

Showcasing the Remarkability

The work was presented at Applied Machine Learning Days 2019 at École Polytechnique Fédérale de Lausanne (EPFL), Switzerland - the global innovation-leading hub. Applied Machine Learning Days is one of the grandest AI and machine learning events in Europe. The event lays its foremost emphasis on the applications of machine learning and AI, metamorphosing them into interest areas for the industry, academia, and public organizations. Along with this, Chaturvedi later also presented this extraordinary endeavor at the Open Data Science Conference (ODSC) East 2019 in Boston and the Global Big Data Conference 2019 in California.

The stunning revelations of Chaturvedi's study accompanied by data science and technology, right from the tobacco marketing methods creating a hazardous urban environment to how the tobacco manufacturers have long targeted minorities and at-risk populations, can assist in the alleviation of the global health efforts.