Oral cancer screening app

Evolution of the Oral Cancer Screening App

Weather app image
Role

Product designer

Platform

Mobile app

Tool stack

Figma, Dropbox, Google docs

Project Duration

12 weeks

Project Background

  • AI-COE (Artificial Intelligence - Center of Excellence) is an initiative from the Indian government that is a path to introduce digital products to the human good using AI's capabilities. One of the product is Aarogya Aarohan

  • The project was sponsored by IISc (Indian Institute of Science), Niti Aayog, and governed by ICMR (Indian Council of Medical Research)

Current Problem

"India has one of the highest oral cancer burdens in the world, with an estimated 77,000 new cases and 52,000 deaths annually, 

The incidence is particularly high among men, and the risk increases with age, particularly in individuals over 40."

Some reasons behind it

Late Diagnosis: 

Many cases of oral cancer in India are diagnosed at an advanced stage due to limited access to health care, low awareness, and reluctance to seek medical help early. Advanced-stage cancer is more difficult to treat and has a lower survival rate.

Inadequate Healthcare Infrastructure: 

Rural areas, where most of the population resides, often lack adequate healthcare facilities. This leads to delayed diagnosis and treatment. Even in urban areas, there can be a shortage of specialized care for oral cancer.

There are other reasons as well, like Low awareness and education, HPV, Poor oral Hygiene, and Widespread use of tobacco

The measure issue discovered by the IISc is Not getting the early detection of oral cancer. In order to tackle that, IISc managed camps in rural and remote areas to spread awareness about oral cancer

This is where Aarogya Aarohan was born

My responsibility and Roles

The project was handed to me after the first phase release until that time team had done pilot testing in Delhi (AIIMS) with Aasha health workers and nurses.

  • The research team has come up with the issues found in the usability testing in Delhi. I was involved in the brainstorming session to solve those problems and introduce new solutions to the app.

  • I was responsible for conducting research on-site in the upcoming releases and integrating the required changes for the next 2 releases.

  • Modules I have to cover - Usability research, User interface design, Improving user experience, and Creating a Design System.

Phase 1 Release

Analysis Based on Usability in Delhi (Phase 1 insights)

"we need draft"

why draft is needed?

  • The Asha workers use their own phones and internet(4g/5g).

  • No electricity is available on the site.

  • Low-end phones, Coasting around 6k to 15k, mostly Android

  • Low mah battery phone which gives them 3-4 hrs of backup

  • Insufficient for multi-tasking.

  • If they get a call on the sim or have to do another operation, the app fetches an error and they have to restart the screening process.

We need to segregate Un-synced cases

Since this app follows an offline-first approach, users are having difficulty distinguishing between synced and unsynced cases. We need to separate these cases so that users can easily see how many are yet to be synced.

Earlier Integration of drafts

  • I have to come up with an idea to integrate the draft that fits into the current design.

  • Also, for the unsynced cases, User must able to scan it quickly and perform an action towards sync button

Wireframes

I started with the layout and came up with some options, The layout options I presented to the team were considered based on usability as well as the requirement from the tech team, We were working on a fast-paced environment, so I quickly created base wireframes as well

Hi-Fi of the second phase

After a bunch of reviews, I created the hi-fi screens and sent them to the tech team with a clear set of guidelines. Also had multiple meetings with the tech team on how the draft and offline first approach will work.

Following is the flow that was finalized

In this phase releases, I was focusing on segregation of the each case type, So I separated draft and navigation in each tab and added manual sync button at the top app bar. I was aiming to show everything upfront and avoid complex navigations inside the app.

2nd Usability study @krishnagiri

The Phase 1 usability study served as a pilot test, providing us with a starting point. To gather more insights from rural and remote areas, we have planned additional usability studies in collaboration with the IISc team.


In Krishnagiri, the IISc team organized two camps in rural areas with the support of the Tamil Nadu government—one in Mallapadi and the other in Kanthikuppam.


I was assigned to Mallapadi, working alongside one Product Designer from Triveous and one IISc researcher.

Context of backend

before moving forward I want to give a bit of context on how Aarogya Aarohan will work in a back-end

key points to consider before moving forward-
  • We can not consider that AI will be fully accurate. As it can’t be more sensitive with its decisions we have to integrate human interventions.

  • In the First 2 phases, the app will be used for training and testing purposes. We needed to integrate the follow-up feature, Which will give the final answer (as humans are taking time to analyze the images)

Phase 3 decisions

Now I have to improve the required changes that come after usability and also Integrate the necessary features.

- Profile

- Notifications

- Follow-up (Recommendations)

- Dashboard

Navigation

To include all these features in the app I needed to redesign the entire app because the current layout did not support all of those features
So I started from scratch meaning with the information architecture and site map

After multiple self-brainstorming sessions I finalised the IA and sitemaps, Then I started with sketches and Lo-fi wireframes.

I created the wireframes to decide the layout also in between we had daily huddles with the IISc team as they were conducting more research on app usability in the Banglore site.

based on those inputs I converted the screens into Hi-fi.

As we earlier leveraging the design system from the simple app. We decided to create our own design system, using the base of a material design system. I created all the components for the mobile app. and for the backend, we continued to carbon design system as we were using it for Midas

Final screens

  • Dashboard screen showcasing stats of screened cases and followup tasks

    Weather app image
  • Cases screen, showcasing glimpse of screened cases, drafts and recently submitted section.

    Weather app image
  • Tasks section showcasing sections of newly added tasks, pending and completed.

    Weather app image
  • Dashboard screen showcasing stats of screened cases and followup tasks

    Weather app image
  • Cases screen, showcasing glimpse of screened cases, drafts and recently submitted section.

    Weather app image
  • Tasks section showcasing sections of newly added tasks, pending and completed.

    Weather app image

Future goals

Currently, phase 2 is released in 5 states, which are Karnataka, Tamilnadu, Delhi, Assam, and Uttar Pradesh

We are conducting more research, To gain valuable insights from our primary users

Conclusion

The AI-Powered Oral Screening App is a game-changer in the fight against oral cancer. By combining AI with a user-friendly design, it empowers healthcare workers to provide better care and improve patient outcomes. This project shows how technology can transform healthcare delivery, especially in underserved communities.

The app is live in Karnataka and Tamilnadu, used by 100’s of Asha workers, Would love to give in detail journey of making AI-COE (AI-Center of Excellence)

Follow me on other channels

hey@mangeshux.in

+91 8483055040

© Mangeshux.in

Follow me on other channels

hey@mangeshux.in

+91 8483055040

© Mangeshux.in

Follow me on other channels

hey@mangeshux.in

+91 8483055040

© Mangeshux.in

What is aarogya Aarohan?

  • Aarogya Aarohan is a mobile app that helps detect oral cancer by taking images of a patient's mouth.

  • Aarogya Aarohan will set up a new process of oral cancer screening, which will foster the work of FLW (Frontline Workers)

  • Inferencing results with the help of AI.

  • It will instantly generate results with the help of AI. To know more about and it how it will be done, we have created medical imaging datasets for India (MIDAS) checkout the Midas case study as well.