Ahmedabad, Gujarat | October 12, 2025 — Nikunj Vaghasiya, a Minor student in Artificial Intelligence & Data Science (CCE) at IIT Mandi, and Founder of Softcodex, has developed SpectraOral—an AI-powered tool designed to assist in early identification of oral lesions and abnormalities using deep learning and image-based analysis.
Oral cancer remains a major public health concern in India, where early screening can significantly improve outcomes. SpectraOral aims to support dentists, medical students, and healthcare workers by enabling quicker preliminary screening through a simple image upload workflow—especially valuable in settings where specialist access may be limited.
What SpectraOral is (and what it isn’t)
SpectraOral is positioned as an assistive screening tool—not a diagnosis engine. The system analyzes oral cavity images to flag patterns that may indicate potential precancerous changes, helping medical professionals prioritize attention and follow-up.
“SpectraOral is designed to assist medical professionals, not replace them,” said Nikunj Vaghasiya.
“Our vision is to strengthen awareness around early detection and make AI more accessible for public health.”
How it works
The tool uses a machine learning pipeline trained on oral lesion datasets to recognize visual features associated with common abnormalities. The goal is to provide fast, consistent, and explainable screening support—especially in rural and semi-urban environments where screening gaps remain.
Why this matters
India’s healthcare challenge is not only treatment—it’s timely detection. Tools like SpectraOral reflect a growing wave of youth-led innovation where AI is being built for social impact, preventive healthcare, and medical education.
Live Demo
The live demo is available here:
https://aiforall.me/spectraoral/
nv sir
Editorial Note (Important)
SpectraOral is an experimental/assistive AI tool and does not provide medical diagnosis. Any suspected condition must be confirmed by a qualified medical professional using clinical evaluation and standard diagnostic procedures.
