In the rapidly evolving landscape of medical technology, a significant breakthrough has emerged in the fight against metabolic dysfunction-associated steatotic liver disease (MASLD). A recent study conducted by health-tech innovator Briya has demonstrated that the strategic application of artificial intelligence (AI) can fundamentally alter how medical institutions identify and track chronic conditions. By leveraging its proprietary AIRE platform, Briya successfully expanded a clinical study population by over 360%, uncovering thousands of at-risk patients whose conditions were previously obscured within the labyrinth of unstructured medical reporting.
This development arrives at a critical juncture for the healthcare sector. As AI integration accelerates across hospitals globally—with the healthcare AI market projected to reach a staggering $57.4 billion valuation by 2029 according to GlobalData—the ability to turn "dark data" into actionable clinical insights is becoming a cornerstone of modern diagnostic medicine.
The Core Challenge: The "Invisible" Patient Population
MASLD, formerly known as non-alcoholic fatty liver disease, is a pervasive condition that often progresses silently. Because early-stage symptoms are frequently non-specific or entirely absent, patients often go undiagnosed until the disease has reached advanced, irreversible stages of fibrosis or cirrhosis.
The central problem addressed by the Briya study lies in the nature of modern medical records. While electronic health records (EHRs) are the standard for data storage, they often rely on structured diagnostic codes—the formal labels applied to a patient’s file. However, much of the diagnostic gold in medicine is hidden in "unstructured" data: the free-text summaries dictated by radiologists after an ultrasound. These reports, often composed in a stream-of-consciousness format, contain vital clinical observations that rarely make it into the patient’s structured medical coding. Consequently, physicians often remain unaware of critical findings buried in reports generated for unrelated imaging procedures.
Chronology: A Multi-Year Deep Dive
To test the efficacy of its AIRE platform, Briya undertook a rigorous longitudinal analysis covering a four-year window from 2020 to 2024.
- Data Aggregation: The team compiled a massive dataset consisting of 28,795 abdominal ultrasound reports.
- Initial Cohort Identification: Using standard EHR diagnostic coding alone, researchers identified an initial pool of 1,122 patients with diagnosed steatotic liver disease.
- The AIRE Intervention: Briya deployed its AIRE AI and natural language processing (NLP) algorithms to scan the 20,422 unique patient records associated with the ultrasound reports.
- The Discovery: The platform identified an additional 4,036 patients with signs of a steatotic liver that had not been flagged by traditional coding systems.
- Resulting Population: The study population expanded from a modest 1,122 individuals to a robust 5,158 patients, representing a 360% increase in total identified cases.
Supporting Data: The 78% Gap
The disparity between structured and unstructured data highlighted in the study is stark. According to Briya’s findings, only 22% of the patients identified by the AI platform had a corresponding diagnosis documented within their structured EHR. This means that a staggering 78% of the patients identified as having a steatotic liver were essentially "invisible" to standard electronic data retrieval methods.
Without the intervention of the AIRE platform, these 4,036 patients would have remained unrecognized, missing the window for early intervention and preventative care. Furthermore, by integrating this new, larger cohort with existing laboratory data, the researchers were able to calculate the Fibrosis-4 (FIB-4) index for the patients. The analysis successfully identified 4,368 patients who possessed a FIB-4 score above the conventional at-risk threshold, providing clinicians with a prioritized list of individuals who require immediate follow-up care.
Official Perspectives: Bridging the Gap Between Tech and Care
The impact of this technology on clinical practice was best summarized by Dr. Gadi Lalazar, the study lead and head of the liver unit at Israel’s Shaare Zedek Medical Center.
"Briya AIRE fundamentally changed how we were able to identify patients for this study," Dr. Lalazar stated. "Many of these patients may not have known they had steatotic liver disease because the finding was buried in an ultrasound report performed for some other reason. Identifying them earlier creates an opportunity for physicians to initiate treatment before the disease progresses to more serious and potentially irreversible stages."
Dr. Or Shaked, director of medical research solutions at Briya, emphasized that the goal is not merely to find more patients, but to empower medical professionals to focus on the human side of medicine rather than the administrative side of data management.
"AIRE’s specialised research agents were used to extract information from unstructured fields, harmonise it with existing structured data, and build the study population," Dr. Shaked explained. "The goal is to take on more of the complex data preparation work so researchers can use more of the available information and focus on answering scientific questions that can ultimately influence care."
Implications: The Future of Preventive Medicine
The implications of this research extend far beyond the specific treatment of MASLD. The success of the Briya project serves as a proof-of-concept for the wider application of NLP in hospital systems globally.
1. Enhanced Research Capabilities
By unlocking unstructured data, medical institutions can conduct retrospective studies with much larger, more representative patient populations. This increased data density allows for more accurate clinical modeling without the need for costly and time-consuming manual chart reviews.
2. Streamlining Clinical Workflows
The automation of data extraction significantly reduces the burden on clinical research staff. As AI handles the heavy lifting of identifying patient cohorts, medical professionals can pivot toward patient outreach and therapeutic planning.
3. Economic and Societal Benefits
The early identification of chronic diseases like MASLD has massive long-term economic implications. By shifting care from late-stage treatment—which often involves liver transplants or intensive chronic disease management—to early-stage lifestyle and pharmacological intervention, the healthcare system can potentially save billions in long-term expenditures while significantly improving patient quality of life.
4. A Shift in Radiology Practice
As AI tools become more integrated, we may see a transition in how radiologists dictate reports. While "stream-of-consciousness" reporting has been the norm, the knowledge that AI will be parsing these texts for diagnostic indicators may lead to more standardized, yet still comprehensive, reporting practices.
Conclusion: The AI-Driven Healthcare Horizon
The collaboration between Briya and Shaare Zedek Medical Center illustrates the transformative potential of artificial intelligence in the medical imaging space. By converting the vast, untapped reservoirs of unstructured radiology data into structured, actionable intelligence, healthcare providers are gaining the ability to identify, risk-stratify, and treat patients with unprecedented efficiency.
As the industry moves toward 2029, the trajectory of AI in healthcare remains clear. We are moving away from an era where diagnosis was limited by the manual search of records, toward a future where data works for the clinician. For the thousands of patients whose liver disease was previously "buried" in a report, this technological leap is not just a triumph of data science—it is a vital, life-altering intervention.
Briya has announced that it plans to present these findings at upcoming medical meetings, where it is expected to generate significant discussion regarding the standardization of NLP-based diagnostic tools in clinical settings. As other institutions look to replicate these results, the standard of care for chronic liver disease, and perhaps other conditions hidden within radiology files, is poised for a significant and welcome evolution.
About the Author
