Imagine Peter. He is a 67yr old male living in Johannesburg, South Africa. He has been on the same treatment for HIV and hypertension for the past 2 years and his medication has not changed in this time. On the surface, he is healthy.

He is due for his 6-monthly follow-up at the clinic soon and he is scared, as he will need to take 2 taxis to get there and not be able to practice appropriate social distancing throughout, making him vulnerable to the virus due to his age.

While also given that the visit is unlikely to change the management of his diseases. We realise the indispensable need to stay at homes to curb the spread of COVID 19 , but more importantly look after our health .

So we decided to create TextCollect(our project) to concurrently meet both the requirements of social distancing plus healthcare.

What it does

TextCollect triages patients (accurately and with reliability) with chronic illnesses using a USSD model in which a patient answers a questionnaire similar to the one used by a doctor and depending on his responses which are relayed to our server to match with the machine learning data, we notify him whether his medication is sufficient for the disease or a change in dosage/medicine is needed. This helps them access health care facilities while maintaining social distancing norms.

Technical Functioning

1)USSD session is initialised by mobile user

2)Sends: HTTP “GET” Message to 3rd Party Server Address

3)XML Response String Containing Menus etc get relayed from the server

4)USSD Menus Displayed on Mobile Handset

5)Responses given by user relayed back to our server for implementing with machine kerning data sets

6)The result of evaluation is provided to the customer via SMS.

We have done extensive case research for this model in South Africa, where majority doesn't have access to smartphones and using USSD will be most effective.

We have a team of data scientists, developers, health care professionals working on this and are ready with the prototype keen on implementing it for the greater good.

Accomplishments that we're proud of

We are proud to have managed all the technicalities and logistics with minimal resources and to have progressed by leaps within minimal time as well.

We are registered as a Delaware Corporation and in the process for patenting the idea.

Extremely proud of our well estimated statistical prediction of this USSD model preventing 8,00,000 visits annually with a mere 10% implementation, implying a massive curb on the spread of COVID-19 and patients peacefully being able to access healthcare.


We aim to breakeven in Year 1 and have plans for payback in Year 2 as explained in the financials document,

Our conservative projection is based on a gradual adoption of our solution with a start at 2% of the target population. Our revenue system is based on a monthly subscription of $0.5, which can be financed by an NGO.

Our system is profitable from the first year :).

What's next for TextCollect

Target the South African govt. for medical response budget, look out for local NGOs like the Praekelt Foundation and foreign agencies like PEPFAR, DIFD.

Building partnerships with private pharmaceutical companies as 26% of all South Africans take at least 1 medication regularly - i.e 15.8 million for HIV alone Among those most at risk for complications of COVID-19, those above 65yrs, amount to 59% of total patients that take at least 1 medication. But 71% of all medications are prescribed in the public sector, and among those accessing health care in public sector, 20% reported of not being able to fill a script in the past year due to stockouts at their clinic. It is that gap we are hoping to fill.

Aim for global expansion of this model much needed in countries with dense populations.

Looking forward to a happy peaceful world enabled with contact List Interaction and minimal travel maintaining social distancing.

Share this project: