There are over 80,000 cases of skin cancer are diagnosed in Canada each year. There are more new cases of skin cancer each year than the number of breast, prostate, lung and colon cancers COMBINED. Canadian born in the 1990s have 2-3 times higher risk of getting skin cancer in their lifetimes (1 in 6) than those born in the 1960s (1 in 20). Skin cancer is not only one of the most common types of cancer, it is also one of the most preventable, detectable, and treatable. Knowing this, we invented Candetech!

What it does

Candetech works by using Google Cloud API to compare an image of a skin lesion against a database of benign skin cancers and malignant skin cancers. It then determines whether that particular image is BENIGN or MALIGNANT based on machine learning. In addition, we also used We also created a web-prototype that would mimic the app. We also created a ML using python and various libraries that predicts malignancy of cancer based on different attributes (e.g., size, radius, etc.). We were able to build a proto-type of the app as well, to show how patients and doctors can both benefit from this app.

How I built it

Figma was used to build the prototype of the app Google Cloud Auto AML was used to train custom made machine learning models using images Python and various libraries was used to make a ML system based off different attributes of cancer CSS/Javascript/Angular/HTML/full mean stack was used to build a proto-type of the website

Challenges I ran into

Training the images took a really look time in Google Cloud Learning how the different libraries work in Python was challenging Trying to integrate node.js app with the Google Vision API Trying to design a user friendly interface

Accomplishments that I'm proud of

Building a finished product with a group of people we just met

What I learned

How to use FIGMA, how to use ML, how to make a perfect pitch

What's next for Candetech

Build functional apps

Share this project: