The problem WOMANHOOD solves

There is a lot of fear of the stigma associated with reproductive health services, women often avoid making use of such services. This stigma imposes a great deal of mental stress, fear, and depression on patients and causes delays in the diagnosis and treatment of their conditions. The importance of women health is not just very low on priority but overlooked at a level that there is a lot of delay for diagnosis of even severe health issues such as PCOS and breast cancer, leave aside loads of generic diseases. Women should never feel ashamed about their health concerns whether that be their periods, reproductive health, menopause or anything else.

Our solution:

Through Womanhood, we aim at solving this barrier for women to connect with doctor and try to reduce the errors and hassle in the diagnosis of the ailments. Womanhood is an initiative through which we want to build an all-in-one women health platform, which diagnoses a lot of issues, especially women specific ones and help raise awareness for its treatment and management. The details have been given in the features.

Features:

Predict PCOS using AI in a person on the basis of their lifestyle and history.
Convey general information about diseases that women are prone to and importance of women healthcare
Run a general diagnosis for the user based on the symptoms she enters.
Breast Cancer Detection tool : Detects the presence of Metastatic Tissue and Invasive Ductal Carcinoma using two AI Models.
Dedicated Dashboard for Patient and Doctor
Tracking of previous Prescription & Medical History
Menstrual Cycle Tracker
Booking Lab/Appointment
Video Conferencing for expert consultancy
Integrating the Machine Learning models in the web application was the main challenge we faced. Extensive research, mentorship from seniors in hackdata and medium blogs helped us get through the bugs we were facing.

Challenges we ran into

Due to bias towards negative values in medical data, accuracy can be a bad metric as it gives an impression of a good performing model but in fact is biased, and won't perform well with other metrics. Used other metrics like precision score.

Choosing the best algorithm and various hypertune parameters for the same. We tested quite some models before finalising the one.