The collaboration between Fundació Puigvert and IOMED identified 50 high-likelihood Alport Syndrome patients using AI and natural language processing (NLP) applied to unstructured clinical records. This approach made patient identification three times faster than traditional manual methods.
context
Chronic Kidney Disease (CKD) affects millions worldwide and can often hide underlying hereditary conditions. Around 10–30% of CKD cases are caused by Hereditary Kidney Diseases (HKDs), yet many remain undiagnosed due to the complexity of symptoms and how they’re documented in medical records.
This study, led by Hospital Fundació Puigvert in collaboration with IOMED, harnessed the power of real-world data (RWD) and artificial intelligence (AI) to support earlier detection of HKDs like Alport Syndrome. By helping physicians find patients whose conditions might otherwise go unnoticed, the study opens the door to faster diagnosis and better care.
the challenge
Alport Syndrome is a hereditary kidney disease that often goes undetected due to its varied symptoms and how they’re recorded in Electronic Medical Records (EMRs). Key signs—like hematuria, proteinuria, or hearing loss—are typically hidden in free-text notes, making them hard to analyze at scale. Traditional identification methods are slow and rely on manual review, limiting timely diagnosis and delaying access to treatment.
The main objective was to find patients with CKD who might actually have an undiagnosed HKD, and support their diagnostic journey using IOMED Data Space Platform.
how iomed helped
IOMED implemented its AI-powered Data Space Platform that extracts clinical concepts from unstructured medical texts using NLP and structuring it through the OMOP Common Data Model.t
The platform focused on detecting patients with phenotypic features linked to Alport Syndrome, a well-defined hereditary condition. Using advanced machine learning methods, we uncovered 50 high-likelihood patients for specialist review and genetic testing. This approach not only made the identification process faster, but also more precise and scalable across institutions.
Results
50
Patients identified as potential Alport Syndrome cases through AI-powered Data Space Platform.
3x
Faster patient identification compared to manual review methods, significantly reducing time to diagnosis.
45K+
Clinical records processed to detect relevant symptoms hidden in unstructured medical texts.
The innovative and advanced technology that IOMED is developing has the potential to transform healthcare, making it more efficient and accessible.
Dr. Mats Sundgren
IHD European Institute
IOMED’s innovative technology offers a robust solution for unlocking the potential of unstructured clinical data, streamlining patient recruitment, and accelerating groundbreaking clinical research.
Dr. Martin Ingvar
Karolinska Institute
The collaboration with IOMED has allowed us to identify and diagnose patients who previously lacked a precise diagnosis, while ensuring data protection at all times, with all data remaining within the institution.
.png)
.png)
.png)
