Docarion

Clinical Intelligence & Telehealth for frontline healthcare across Africa. Connecting clinics, pharmacies, labs & hospitals. Launching soon 🩺

[Docarion @docarion_hq]

Your next doctor visit will be just down the street, and probably right form you room if you choose. Docarion is coming soon to your neighborhood and right into your homes, just like in the days of early medicine. 🇳🇬

[Ayilola of Nigeria /Laptop Guy 🩺 🇳🇬 🇬🇧🇺🇸 @silverpenydr]

When you fall sick at home in Nigeria, your only options shouldn’t be a 5-hour hospital queue or self-medicating at a local chemist with no medical training. 🇳🇬

For months, @evidencechecked, @Doctware @KunnujiC and I have been building @docarion_hq to get you in front of a licensed doctor in under 10 minutes:

⏱️ 1 min: Set up your account once for life
💬 4 mins: Quick AI symptom check-in (English, Pidgin, Hausa, Yoruba, Igbo)
👨‍⚕️ < 5 mins: Consult a real doctor & get your prescription across 200,000+ care points

We’re representing in the @CerebralValley × @AIatMeta Global AI Hackathon (with @basetenco & @modal_labs)!

Every ❤️, Repost & comment helps us bring this home 👇🚀

[Docarion @docarion_hq]

@SirDavidBent Thank you Chief @SirDavidBent!

[Docarion @docarion_hq]

@osemagnum Appreciate you deeply Dr. Ose @osemagnum! 🙏🏾🩺 Built by clinicians for clinicians and patients—can't wait to have you experience @docarion_hq firsthand!

[Dr. Ose Etiobhio @osemagnum]

This is impressive my bro. 👏👏👏👏

[Docarion @docarion_hq]

@OurFavOnlineDoc Thank you so much @OurFavOnlineDoc! 🙌🏾🩺 Empowering patients to speak to a licensed physician from home or before buying medication at the —with a lifelong longitudinal health record—is how we change healthcare access at scale. Follow @docarion_hq!

[Docarion @docarion_hq]

@Big_khaddyy @silverpenydr @evidencechecked @Doctware Thank you so much @Big_khaddyy! Follow @docarion_hq to ride with us!

[Docarion @docarion_hq]

@The_Real_EJC @evidencechecked @silverpenydr @Doctware He's hat we believe, when checking a symptom feels as normal as checking the time or taking an evening outdoor walk, prevention stops being a campaign and becomes the default. And yes, we have how possible it is.

[Docarion @docarion_hq]

@musebooklol @evidencechecked @AIatMeta Appreciate that so much @musebooklol!

[Docarion @docarion_hq]

Spot on @AgbaBlog1! AI handles the heavy lifting—maintaining the patient's longitudinal history, local-language intake, and asssited SOAP notes which is 60-70% of the entire consultation job—so our licensed doctors can focus 100% on clinical judgment and safe prescriptions. Appreciate the support—follow @docarion_hq! 🙌🏾

[Docarion @docarion_hq]

@JapanAsItIs @silverpenydr @evidencechecked @Doctware You nailed the exact heart of why we built @docarion_hq! Instead of fighting people's natural habit of walking into a chemist when unwell, we put a licensed doctor + a lifelong longitudinal health record right inside that moment before they buy medication. Thank you!

[Docarion @docarion_hq]

@NotLikeTokz @silverpenydr @evidencechecked @Doctware teh scanner in our system empowers you as end user to confirm the authenticity of every pack you're handed. empower the people and the system will work.

[Docarion @docarion_hq]

@vicsonham1 @silverpenydr @evidencechecked @Doctware Its designed so even someone’s grandmother can use it without typing a word! Usual habits, innovative systems that connects to a licensed doctor in minutes. Follow @docarion_hq as we roll out!

[Docarion @docarion_hq]

Incredible thread @raedle! 🙌🏾 We launched our hackathon build yesterday —putting SAM 3.1 + Muse Spark + Muse Wonder to work in frontline healthcare in Nigeria 🇳🇬🩺

Docarion uses SAM 3.1 (on Modal GPU) to segment RDT strips, measure chronic wounds (cm²), redact patient PII on paper labs before OCR, and verify NAFDAC drug packaging—helping patients talk to a real doctor before buying medicine at 200,000+ pharmacies:
https://t.co/aU7Lzu0fwN

[Ayilola of Nigeria /Laptop Guy 🩺 🇳🇬 🇬🇧🇺🇸 @silverpenydr]

When you fall sick at home in Nigeria, your only options shouldn’t be a 5-hour hospital queue or self-medicating at a local chemist with no medical training. 🇳🇬

For months, @evidencechecked, @Doctware @KunnujiC and I have been building @docarion_hq to get you in front of a licensed doctor in under 10 minutes:

⏱️ 1 min: Set up your account once for life
💬 4 mins: Quick AI symptom check-in (English, Pidgin, Hausa, Yoruba, Igbo)
👨‍⚕️ < 5 mins: Consult a real doctor & get your prescription across 200,000+ care points

We’re representing in the @CerebralValley × @AIatMeta Global AI Hackathon (with @basetenco & @modal_labs)!

Every ❤️, Repost & comment helps us bring this home 👇🚀

[OurFaveOnlineDoc 🇬🇧 🇳🇬 @OurFavOnlineDoc]

This is another good way to get patients to speak to a doctor if you need medical help. If you fall sick at home and it’s difficult to see a doctor, this is a good initiative

[Docarion @docarion_hq]

@Wallevic @silverpenydr @evidencechecked @Doctware This took almost 1 year!

[Docarion @docarion_hq]

Great question!
1. Our clinical vision and voice models are self-hosted on our own isolated GPU infrastructure, and every prompt passes through a deterministic PHI-scrubber firewall before inference.
2. Identifiable patient data is never used to retrain or fine-tune foundation models, and is never shared with third-party model providers. All records are encrypted and strictly governed under NDPA 2023 privacy standards.
3. With regards to training models, our triage engine, red-flag detectors, and differential diagnosis (DDx) models are built on an advanced clinical repository tailored to Nigerian disease states, local clinical guidelines, and real-world longitudinal patient records rather than generic symptom trees.

[Docarion @docarion_hq]

For the second and third layer, we had a pre-claim triage through dispensing, this enforces end-to-end medication safety by surfacing red flags, prior dispense and active allergies in the pre-claim Physician Brief, cross-checking treatments in real time check via a continous Co-Pilot and e-Rx writer. By doing this, we locked pharmacy dispensing behind a mandatory four-point clinical and regulatory batch safety check.

I had no way of expailning this to engineers, i had to literally lay the bricks brick by bricks for this.

[Famotire Akinwale V. @Wallevic]

@silverpenydr @evidencechecked @Doctware @docarion_hq Great work by your team! I’m very curious, how did you handle the medical text similarity considering the fact that our local languages have low representation in clinical NLP data? Also, how do the doctors get the patients’ medical history to check for contraindications?

[Docarion @docarion_hq]

@Wallevic @silverpenydr @evidencechecked @Doctware And yes, for the last part of your question, still details i cant provide here but we enforce a three-layer safety architecture so doctors and pharmacists never prescribe blind, with teh first layer ben a Standardized Fixed Longitudinal History.

[Docarion @docarion_hq]

with regards to medical text similarity, rather than relying on raw cosine similarity over low-resource language embeddings—which fail when patients describe symptoms colloquially—we use a two-stage Clinical Concept Normalization + Hybrid Similarity pipeline. Details not for this space.

[Docarion @docarion_hq]

@thewitnez @silverpenydr @evidencechecked @Doctware Your place in line and appointment is held for 2hrs.