Most people these days tend to search about their personal problems on the internet. Thus, we believe utilizing people’s search entries can be a great starting point to detect whether they would need mental healthcare support. Our browser extension analyzes users’ search inputs from the browser’s address bar through a minimum of two weeks period and gives a negativity score to each individual. Individuals who score high on the negativity chart will be prompted by our extension to talk to our chatbot application backed by an AI that is trained with data from the Beck Depression Inventory. The chatbot will calculate a depression score that will provide the first pre-diagnostic data to the general practitioner, provide a self-help guideline to the high-risk individuals. This way, general practitioners will easily know if the patient needs to be referred to a mental health professional as well as differentiate psychological and somatic symptoms. Bastion aims to breeze the individuals with depression into a mental health specialist as quickly as possible to lower undiagnosed depression rates.
We plan to develop our project into a flexible API that can be implemented both by local healthcare tracking systems such as EESZT in Hungary and by other organizations operating on a global scale. We are also planning to add features that can inform non-professionals on the symptoms and treatment methods of depression so that people living in a fragile setting such as war, scarcity, etc. can help their community members tackle depression. Bastion can also provide a crucial database to healthcare researchers by showing local trends that might lead to depression and various other mental health problems all while protecting user privacy through data through pre-trained AI datasets which ensures no personally-identifiable user data ever leaves their device.
