AI-Powered App Analyzes Fingernails to Identify Nutrient Deficiencies

A high school student developed an Artificial Intelligence (AI) app to identify nutrient deficiencies. It will also suggest dietary changes after the diagnosis.

AI-Powered App to Identify Nutrient Deficiencies

In 2020, Rian Tiwarihas, a high school student at South Brunswick High School in New Jersey, was learning remotely due to the COVID-19 pandemic. He found himself bored, and his father suggested thinking about ikagai, a Japanese concept that gives a person a sense of purpose.

He decided to help others through technology. He researched chronic conditions. He discovered that there was already an app to monitor hemoglobin levels and learned that a fingernail's appearance is key to a person's health. He focused on anemia.

Changes in the appearance of a nail, such as discoloration, ripples, lumps, and other variations, might be indicators of nutrient deficiencies or diseases. For instance, white spots on the skin could be an indication of a zinc deficiency. A lack of folic acid can be seen in the form of brittle and cracked nails.

Tiwari developed an application that examines fingernail scans for indications of shortages in various minerals, including vitamin B12, calcium, and zinc. If any of these warning indicators are present, the mobile application will suggest dietary and lifestyle adjustments that could stop the progression of anemia.

What is Anemia?

Anemia is when the body doesn't have enough healthy red blood cells, which provide oxygen to body tissues, according to Penn Medicine. One can suffer from anemia for several reasons, such as follows:

• Anemia due to vitamin B12 deficiency
• Anemia due to folate (folic acid) deficiency
• Anemia due to iron deficiency
• Anemia of chronic disease
• Hemolytic anemia
• Idiopathic aplastic anemia
• Megaloblastic anemia
• Pernicious anemia
• Sickle cell anemia

How Did AI Read Fingernails to Identify Nutrient Deficiencies

Tiwari created the algorithms used in the app by running data sets obtained from the open-source website Kaggle through a machine-learning platform.

The app starts with a photo of a nail. It will then use a device-based neural network to look for cracks, ridges, peeling, and discoloration and classify if the user is healthy or unhealthy.

If the app detects that the owner of the fingernail has a folate deficiency, it will recommend foods like asparagus, spinach, and sunflower seeds.

The app also stores medical information and records of the analytics and recommendations from previous scans.

What's Next For the App?

In 2020 Tiwari and two of his classmates submitted a pitch and a business plan for their product to the Conrad Challenge, a competition for students developing technologies that address a global problem. They were presented as finalists, and they presented the idea virtually to the Conrad Challenge Innovation Summit at the Space Center Houston.

They didn't win, but he moved ahead with the app development. He asked the help of Viswanatha Allugunti, a solutions architect at Arohak.

According to him, Allugunti helped him push himself and the project to the next level. Allugunti taught him more about coding, filing a patent, and how to conduct effective research.

Allugunti also helped him narrow his target users from anyone with anemia to pregnant people. He focused on a certain group of people with particular needs.

He contacted maternal health organizations to test the app. He also spoke with Reach, a nonprofit that provides mentorship and educational opportunities to people developing technology that enables individualized patient care.

Tiwari is working on getting the app to analyze images of lips and inner eyelids, which can also show signs of nutrient deficiencies. He also wanted to expand the features of apps, like adding recommendations for medications and vitamins.

The app will pilot this year, and Tiwari wants to make it available for Android and iOS devices.

Check out more news and information on Medicine and Health in Science Times.