I am thrilled to share that my conference paper, "Lightweight Harmonic Frequency–Spatial Fusion GAN for Macular Hole Image Generation," has been published in the Disrupting for Good: AI, Entrepreneurship, & Sustainable Circular Economy - 2nd Edition conference proceedings!
Abstract: This work introduces the HMDF-GAN-Lite, an innovative, lightweight model designed to generate clinically realistic fundus and OCT images for the early detection of macular holes. With a focus on Generative Adversarial Networks (GANs) and AI-based diagnostic studies in ophthalmology, our model provides a new approach to dataset augmentation and rare case modeling in retinal imaging.
Key Achievements:
Reduced model parameters by half, improving computational efficiency.
Enhanced the ability to generate images reflecting the true pathologic manifestations of macular holes.
Highlighted the potential of AI for sustainable healthcare by leveraging low-cost, high-fidelity image synthesis.
This is just the beginning for AI in healthcare, and I’m excited to contribute to this growing field!
Stay tuned for more updates, and let’s continue the conversation about the power of AI in healthcare innovation! #AI #MachineLearning #Ophthalmology #GenerativeAdversarialNetworks #SustainableHealthcare
Building a Voice AI Assistant That Turns Phone Calls Into Thoughtful Follow‑Through
I built a Django backend for a family-focused AI phone assistant. A registered caller connects through Twilio, and Django Channels streams the call to an OpenAI Realtime session for a natural, two-way conversation.
The backend adds caller-specific context, selected voices, subscription minute limits, and interruption handling. After the call, background tasks generate a separate summary. A follow-up email is sent only when requested and the recipient is confirmed.
I also built APIs for managing relations, subscriptions, and dashboard usage data. One of the most interesting parts was connecting the whole lifecycle: incoming call, real-time conversation, usage tracking, and post-call follow-through.
Sending the follow-up email only after the recipient is confirmed is a careful rule: a wrong name heard on a call could send it to the wrong person. How does the confirmation work: does the assistant repeat the name during the call, or does the caller confirm later in the dashboard?
Just launched Voltaryx - an AI-powered SaaS template built in Framer
When designing for AI platforms and workflow automation tools, the goal is making complex software feel effortless and trustworthy from the first scroll. For Voltaryx, I paired soft gradient aesthetics and crisp typography with conversion-focused layouts so startups can communicate value fast.
Highlights of the build:
✦ Interactive dashboard & analytics previews
✦ Built-in Day & Night Mode toggle 🌗
✦ Integrations showcase & pricing comparison tables
✦ Full CMS support for Changelog & Blog insights
Available now as a ready-to-launch template for founders—or if you’re building a custom SaaS product and need a tailored Framer site from scratch, my DMs are open for new projects!