Detect aging foot conditions with AI, in seconds.
AFCDAI identifies common foot conditions - fungal toenail infections, corns, calluses, and long toenails - from simple photos. A privacy-first clinical decision support tool, trained on six years of real nursing data and validated by regulated foot care professionals.
Three taps from photo to foot-health insight
No appointment, no paperwork. AFCDAI turns simple foot photos into clear, clinically informed guidance - and connects users to a foot care nurse when it matters.
Capture
Guided prompts help capture the soles, close-up toenails, heels, and sides of both feet - no special equipment required.
Analyze
Images are anonymized, background-removed, and run through condition-specific AI classifiers in seconds on secure AWS infrastructure.
Results & care
Get instant, easy-to-read condition insights - and the option to book a regulated foot care nurse to take the next step.
Conditions AFCDAI detects
Subtle, overlapping foot conditions are hard to tell apart - even for trained clinicians. AFCDAI was built and validated to distinguish them, with accuracy measured during clinical validation.
Fungal Toenail Infection
Discoloured, thickened, or brittle nails caused by fungal infection - identified with high accuracy.
Corns
Small, hardened areas of skin from pressure or friction, often on or between the toes.
Calluses
Thickened skin from repeated pressure, commonly across the soles and heels.
Long Toenails
Overgrown nails that increase risk of injury and discomfort - flagged for timely care.
The technology behind AFCDAI
Built through rigorous experimental development - the kind of R&D recognized under Canada's SR&ED program - to make medical-grade image classification reliable on real, imperfect data.
DenseNet121 Architecture
A proven medical-imaging backbone, adapted via transfer learning from ImageNet with custom classification heads.
Per-Condition Classifiers
Separate binary classifiers for each condition reduce inter-class interference and sharpen accuracy.
REMBG Background Removal
Isolating clinically relevant regions removed noise and significantly improved feature discrimination.
MONAI Standardization
A structured medical-imaging pipeline - resizing, normalization, and tensor conversion - for consistent inputs.
Data Augmentation
Rotations, flips, scaling, and intensity adjustments addressed class imbalance and improved generalization.
Secure AWS Cloud
Hosted on Canadian-aligned AWS infrastructure for scalability, high availability, and strong data protection.
No names. No faces. No personal records.
AFCDAI was deliberately built to minimize data exposure - it functions strictly as a clinical decision support tool, never as a personal health record.
No identifiable data
No names, emails, phone numbers, medical history, or payment details are ever collected or stored.
No biometrics or faces
No facial images or biometric identifiers are captured at any stage of the process.
Foot images only
Only anonymized images of feet are processed - solely to identify visible foot conditions.
Not linked to identity
Images are anonymized before processing and never linked to an individual, so no health record is created.
Canadian-aligned cloud
Hosted on AWS using Canadian-aligned infrastructure and security best practices.
Reduced cyber risk
Minimizing data exposure reinforces clinician and patient trust and lowers cybersecurity risk.
Built on real clinical practice, recognized externally
Validated by licensed clinicians
Trained and validated through direct human clinical review by a licensed nurse, with labeling by regulated foot care professionals.
Six years of real-world data
Reflects real clinical practice - not purely academic datasets - using nursing data collected over a six-year period.
Rigorous experimental development
Iterative tuning, confusion-matrix analysis, and validation metrics - with insights into the limits of synthetic medical data.
Smarter foot health,
in your pocket.
Try AFCDAI for yourself, or talk to us about deploying privacy-first foot screening across your clinic, community program, or health organization.
AFCDAI is a clinical decision support tool. Results are not 100% accurate and are not a medical diagnosis - please consult a qualified healthcare professional before acting on any result.
