Loading...
NGO Darpan CSR-1 80G 12A Trust Deed
Competition

Samasya Khoj


Identify Problems. Inspire Innovation.
← Go Back
Problem ID: 1001040
🧠 Problem Title: When Technology Fails to See: Racial Bias in Sensors, Cameras, and Healthcare Devices for Dark Skin
📝 Problem Description

Modern technologies which are perceived to be objective, neutral and universal functional, exhibit significant bias towards darker skin tones. Many widely used devices, including cameras, touchless soap dispensers, facial recognition tools utilize pulse oximeters, i.e they rely on light-based screening mechanisms. These systems are typically trained upon lighter skintones, not reflecting the equality standard.
Due to higher melanin content, darker skin absorbs more light rather than reflecting it, leading to reduced accuracy or complete failure of these sensors. As a result, there can be inaccuracies to detect diseases, or perform medical operations on such people. While this may seem like minor inconvenience, it becomes critical on a larger scale. For instance, pulse oximeters have been shown to provide falsely normal oxygen readings for dark-skinned patients - leading to delayed operations.
Addressing this problem is crucial to to ensure equitable access to accurate, safe and effective technological solutions. Without intentional design, this problem will continue to impact millions of people for decades to come.

⚡ Impact
Hundreds and more
🌍 Impact Area
Universal
📊 Scope
National
📌 Place
Universal
⏱ Occurrence
Frequently
⚠ Life Risk
Yes
💰 Financial Loss
Lakhs or more

👤 Name
Vidya Sharma
🎓 Class
11
📍 State
Delhi