Enterprise AI Analysis: Healthcare HCI
'Socheton': A Culturally Appropriate Al Tool to Support Reproductive Well-being
Sharifa Sultana (University of Illinois Urbana-Champaign, Computer Science, Champaign, Illinois, USA, sharifas@illinois.edu)
Hafsah Mahzabin Chowdhury (University of Illinois Urbana-Champaign, Computer Science, Champaign, Illinois, USA, hafsahc2@illinois.edu)
Zinnat Sultana (S.M.R. Law College, Jashore, Bangladesh, zinnat1409@gmail.com)
Nervo Verdezoto (Cardiff University, School of Computer Science and Informatics, Cardiff, United Kingdom, verdezotodiasn@cardiff.ac.uk)
Reproductive well-being education in the Global South is often challenged as many communities perceive many of its contents as misinformation, misconceptions, and language-inappropriate. Our ten-month-long ethnographic study (n=41) investigated the impact of sociocultural landscape, cultural beliefs, and healthcare infra- structure on Bangladeshi people's access to quality reproductive healthcare and set four design goals: combating misinformation, including culturally appropriate language, professionals' account- able moderation, and promoting users' democratic participation. Building on the model of 'Distributive Justice,' we designed and evaluated 'Socheton,' a culturally appropriate AI-mediated tool for reproductive well-being that includes healthcare professionals, AI- language teachers, and community members to moderate and run the activity-based platform. Our user study (n=28) revealed that only combating misinformation and language inappropriateness may still leave the community with a conservative mob culture and patronize reproductive care-seeking. This guides well-being HCI design toward being culturally appropriate in the context of reproductive justice with sensitive marginalized communities.
Executive Impact: Bridging Gaps in Reproductive Well-being
Our research identifies critical challenges in reproductive well-being education across the Global South, particularly in Bangladesh. 'Socheton' offers a culturally sensitive AI-mediated platform to address misinformation, language barriers, and moderation issues, fostering democratic participation.
0
Ethnographic Study Participants
0
User Study Participants
0
Core Design Goals Addressed
0
Distributive Justice Model
Deep Analysis & Enterprise Applications
Select a topic to dive deeper, then explore the specific findings from the research, rebuilt as interactive, enterprise-focused modules.
Addressing Cultural and Infrastructural Barriers
Our ethnographic study revealed that communities often encounter reproductive well-being content perceived as misinformation or misconceptions, leading to stigma and avoidance of formal healthcare. Traditional beliefs and lack of appropriate language contribute to this challenge. Socheton aims to combat this by providing verifiable information and a platform for democratic content curation.
Addressing Cultural and Infrastructural Barriers
A significant challenge identified was the mismatch between modern Western-centric reproductive well-being language and local cultural sentiments. Many terms used by practitioners or in educational materials were found to be culturally incoherent or inappropriate, hindering effective communication and trust. Socheton integrates AI-language teachers and community input to ensure culturally appropriate language.
Addressing Cultural and Infrastructural Barriers
Users reported difficulties verifying online information and a lack of accountability from content moderators on existing platforms. Facebook's forced moderation policies were seen as insensitive. Socheton addresses this through a multi-layered moderation system involving healthcare professionals, AI-language teachers, and community-elected administrators, ensuring transparency and accountability.
Addressing Cultural and Infrastructural Barriers
Participants expressed a strong desire for their voices to be heard and their cultural sensitivities respected in reproductive well-being content curation. Socheton promotes democratic participation by allowing users to report, challenge, and suggest alternative content, fostering a sense of ownership and relevance within the community.
Socheton's Culturally-Sensitive Content Curation Workflow
User Submits Report (Misinformation / Language)
→
AI Suggests Evaluation
→
Admin (Professional / AI-LT) Reviews & Edits
→
AI Learns from Admin Feedback
→
Culturally Appropriate Content is Updated
Key Engagement Metrics
0 Ethnographic Study ParticipantsKey Engagement Metrics
0 User Study Feedback Participants| Challenge Area | Current Landscape in Global South | Socheton's Solution |
|---|---|---|
| Information Accuracy & Trust |
|
|
| Language & Cultural Sensitivity |
|
|
| User Participation & Empowerment |
|
|
Addressing Inappropriate Content Moderation
The Challenge: Insensitive Content Moderation
In traditional online platforms, community members often face challenges with content moderation that is insensitive to cultural nuances. A participant (P40, Male, FGD-6) shared an experience: 'We have a Facebook group where people generally post about that kind of problem. A group member once posted about their erectile-related problem using local jargon, "Blackwater/Kalpana." However, some members found it inappropriate, and people discussed about allowing this post. Then, the group admin suddenly told us that Facebook had asked them to delete that post as it did not fit the Facebook community standards. Otherwise, Facebook might even close down the group. The admin had to delete it, and the group members were upset.'
Socheton's Solution: Accountable & Culturally Sensitive AI-Mediated Moderation
Socheton addresses this by integrating AI-language teachers and community-elected administrators alongside healthcare professionals. This multi-layered moderation system ensures that reports of inappropriate language or misinformation are reviewed with an understanding of local cultural context and accepted jargon. Users can suggest alternative language, and administrators make informed decisions, allowing the AI to learn and adapt. This approach prevents the 'conservative mob culture' and ensures that sensitive topics are discussed respectfully, fostering trust and democratic participation, rather than suppressing local expressions.
Calculate Your Potential AI Impact
Estimate the efficiency gains and cost savings your enterprise could achieve by implementing a culturally appropriate AI solution.
Annual Cost Savings
Annual Hours Reclaimed
Socheton Implementation Roadmap
A phased approach to integrate culturally appropriate AI, ensuring sustainable and impactful change in reproductive well-being support.
Phase 1: Ethnographic Study & Design Goals (Oct 2021 - Jan 2023)
Conducted a ten-month ethnographic study (n=41) in Bangladeshi villages to understand reproductive well-being challenges and sociocultural landscapes, leading to four core design goals.
Phase 2: Socheton Design & User Evaluation (Jan - Aug 2024)
Designed 'Socheton' as a culturally appropriate AI-mediated tool. Conducted user studies (n=28) to gather feedback on the prototype, revealing challenges and opportunities for design refinement.
Phase 3: Nationwide Language Survey & Model Refinement (Future)
Future work includes conducting a nationwide survey to build a rich database of acceptable local languages and refine the AI model for even greater cultural appropriateness and linguistic accuracy.
Phase 4: Expansion & Stakeholder Integration (Future)
Planned expansion to include LGBTQ+ communities and other marginalized groups. Collaboration with local government, health ministry, NGOs, and religious leaders to ensure broad adoption and impact.
Ready to Transform Your Enterprise with AI?
Discover how culturally appropriate AI can enhance your initiatives and drive meaningful impact. Book a free consultation with our experts today.
Ready to Get Started?
Book Your Free Consultation.
Let's Discuss Your AI Strategy!
Lets Discuss Your Needs
Select a Date
Sun
Mon
Tue
Wed
Thu
Fri
Sat