Brigid

Inspiration

Problem
Breast cancer is a disease in which cells in the breast grow out of control. In 2020, there were 2.3 million women diagnosed with breast cancer and 685 000 deaths globally. As of the end of 2020, there were 7.8 million women alive who were diagnosed with breast cancer in the past 5 years, making it the world's most prevalent cancer. Fast and accurate breast cancer screening has always been an issue for mankind to solve ever since the disease was discovered.
Current screening and prediction methods
Breast cancer prediction methods that have been developed over the years include mammography, breast exam, thermography, and tissue sampling.
Most effective well-known methods
Although mammography is the most widely used method for prediction, it has many downsided including side effects of the mammography itself. Mammography is less likely to find breast tumors in women with dense breast tissue. Because both tumors and dense breast tissue appear white on a mammogram, it can be harder to find a tumor when there is dense breast tissue. Magnetic resonance imaging (MRI) may be used to screen women who have a high risk of breast cancer. A clinical breast exam is an exam of the breast by a doctor or other health professional. He or she will carefully feel the breasts and under the arms for lumps or anything else that seems unusual. It is not known if having clinical breast exams decreases the chance of dying from breast cancer. Thermography is a procedure in which a special camera that senses heat is used to record the temperature of the skin that covers the breasts. Tumors can cause temperature changes that may show up on the thermogram. There have been no randomized clinical trials of thermography to find out how well it detects breast cancer or the harms of the procedure. Breast tissue sampling is taking cells from breast tissue to check under a microscope. Breast tissue sampling as a screening test has not been shown to decrease the risk of dying from breast cancer.
The best method we found
But while we were researching about all of these, we came across a method that analyses fine needle aspirate of a breast mass which provides more accuracy even more so with our model.

What it does

Solution
We developed our own breast cancer prediction model that detects the chances of a person having breast cancer.
Brigid
It is an AI Model to identify if the stage of breast cancer is Malignant (very infectious) or Benign (not that harmful). Brigid is an ancient Irish goddess associated with fertility, the spring season, and healing. FNA biopsy results can then be used to accurately finish the task of predicting cancer.
AI Model for prediction
The model analyses information from a fine needle aspirate of a breast mass, gathers the data, and calculated the probability of a person having breast cancer. The model provides more than 90% accuracy for the tests which makes it one of the best breast cancer prediction models so far.
Chat-bot as smarty-pants
A Chatbot for diagnosis, recommending doctors, and chatting with the user in general. It is like the one friend who always has accurate (or sometimes not), answers to all your questions. We understand the sense of responsibility one takes on along with big things, and hence strive to provide accurate and useful responses.

How we built it

The AI model has been coded in the Jupyter Notebook and uses SciKit Learn and Tensorflow libraries.

Challenges we ran into

Our team consists of mostly first-time hackathon participants so it was a bit challenging to figure out our skillsets and make them all cohesive but we worked hard and produced something we truly consider amazing!