Clinical AI in practice: Sofya at Brazil’s 4th General Medicine Congress
Sofya cofounder and head of AI Bruno Dornelles discussed clinical documentation, adoption, and accountability on the AMB panel.

What we publish, where we present, and how clinical intelligence reaches real care.
Photo · São Paulo Medical AssociationSofya cofounder and head of AI Bruno Dornelles discussed clinical documentation, adoption, and accountability on the AMB panel.
Every item links to its original source. Company claims are identified as Sofya archive; research and recognition link to the institutional or scientific record.
clinical practice01The deployment captures, transcribes, and structures the conversation in real time, with patient consent and final physician validation.
company02Photo · Refresh MiamiThe story follows Sofya’s evolution toward a clinical studio where physicians and health systems can compose their own workflows on secure infrastructure.
At the private executive lunch, Sofya presented how real-time clinical reasoning can make high-value care more precise, scalable, and operational.
company04Photo · Refresh MiamiThe profile presents the idea of a second clinical perspective for physicians, Sofya’s Latin American footprint, and its US growth strategy.
The partnership combines MV’s regional footprint with Sofya’s clinical AI platform to embed documentation and decision support in the EHR.
clinical practice06Photo · Medicina S/AIn his article, Mearim discusses consent, infrastructure, clinical accountability, and the safeguards required when artificial intelligence handles sensitive health data.
partnerships07The cohort work focuses on refining real-time clinical reasoning models for precision, efficiency, and quality of care.
events08At the event connecting the Miami and Dubai ecosystems, Sofya won among ten finalists and received travel and exhibition space at Expand North Star 2025.
events09Frame · Meta Developers / LlamaConOn the stage of Meta’s first global conference dedicated to Llama, Sofya was featured as a clinical AI example that cuts time spent on paperwork during patient visits by 30%.
Sofya was one of three companies selected by peers in the equitable healthcare access program for Latin America.
After an immersion week with founders from Brazil and Uruguay, Sofya was selected as the winner at the Demo Day hosted at Mana Tech headquarters.
company12Image · Times Brasil / CNBCIgor Couto explained to Times Brasil/CNBC how Sofya structures biomedical data and uses clinical intelligence to give physicians time back and support decisions during care.
clinical practice13The case study describes smaller models tuned for clinical tasks, low latency, and up to 30% less administrative time per consultation.
company14Photo · Sofya / IT ForumThe article connects return on investment, clinical productivity, and broader access, arguing for use cases aligned with each healthcare organization’s real needs.
events15At the 2025 kickoff, Igor Couto presented how Sofya listens, understands, and reasons through clinical protocols to support precision care.
The study, with participation from Sofya research, evaluated 969 Portuguese clinical texts and found high sensitivity and specificity for detecting anaphylaxis.
clinical practice17Photo · Medicina S/AMearim examines how integrated data, interoperable health records, and automation can support more personalized decisions without separating innovation, privacy, and accountability.
events18Sofya was selected to present its augmented clinical reasoning approach at the AI and health meeting organized by Stanford AIMI, HAI, and CME.
The paper, with Sofya research participation, evaluated four models and found high sensitivity for identifying true cases in clinical records.
The CBIS paper, co-authored by Sofya research, structured clinical concepts and criteria to flag suspected anaphylaxis with interoperable terminology.
The case showed real-time clinical transcription and analysis, plus the infrastructure required to serve telemedicine operations at scale.
Joice Basílio Machado Marques presented how data becomes knowledge for clinical practice through SNOMED CT, FHIR, and AI.
research23The AAAI Spring Symposium paper structured clinical notes into IPS, clinical impression, and medical encounter data using international standards.
The scientific abstract, with participation from Sofya’s Joice Basílio Machado Marques, presented AI and clinical language processing to support diagnosis.
research25Image · SNOMED InternationalThe international panel on terminologies and AI connected clinical standards, provenance, safety, and the responsible use of generative models in care.
The CBIS 2022 paper, co-authored by Sofya, described an architecture for recognizing and coding diseases, procedures, and symptoms.
The ONTOBRAS paper, co-authored by Sofya, combined SNOMED CT, OWL-Time, and semantic rules to represent risk factors and support clinical classification.
No items in this category.
Reporting and institutional milestones organized as a reading—from clinical documentation to programmable reasoning, validation, and continuous care.
Editorial01of03