Coreline Soft Proves Success in CT-Based Sarcopenia Diagnosis

"Reading Sarcopenia from a Single CT Scan": Medical AI Discovers Hidden Health Indicators in Images 

While muscle loss is a natural part of aging, a decrease beyond a certain level—known as sarcopenia—drastically increases the risk of falls, fractures, and chronic diseases. The World Health Organization (WHO) assigned a disease code to this condition in 2017. Despite its severity and prevalence among the elderly, early detection remains challenging.

Medical Artificial Intelligence (AI) is emerging as a solution. Recent findings confirm that muscle mass, which previously required separate tests or complex measurements, can now be automatically analyzed and used as a diagnostic indicator using existing CT scans.

A study recently published in the international journal Aging Clinical and Experimental Research presented a new standard for diagnosing sarcopenia by automatically analyzing psoas muscle volume from CT images. After analyzing data from approximately 4,000 adults, researchers confirmed that the psoas muscle volume indicator correlates highly with the existing Appendicular Skeletal Muscle Mass (ASM) index, making it a significant metric for diagnosis.

This research highlights the potential of "opportunistic screening," where health risks are discovered early by repurposing existing CT data taken for other purposes—such as general checkups or diagnosing different diseases—without the need for additional tests. Furthermore, since the psoas is a core abdominal muscle, sarcopenia can be diagnosed quickly and accurately by measuring its volume alone, rather than measuring total body muscle mass.

 

"The Answer Was Already in the Images": Medical AI Opens the Era of Opportunistic Screening 

■ Reading Hidden Muscle Information in CT Scans 

On March 26, Professor Ji-wan Kim and Postdoctoral Researcher Woo-rim Choi from Asan Medical Center announced the publication of a paper on diagnosing sarcopenia using Coreline Soft’s 'AVIEW'.

The study evaluated the distribution of psoas muscle volume and its association with sarcopenia by analyzing CT data from 3,999 adults. Using a deep-learning-based algorithm, the researchers automatically segmented the psoas muscle and calculated its volume, comparing it with the traditional ASM index.

The results showed a high correlation between psoas muscle volume and existing indicators. Specifically, the PV/BMI index (psoas volume adjusted by Body Mass Index) showed the highest diagnostic accuracy. Based on this, the researchers proposed a new diagnostic standard for determining sarcopenia via CT imaging. The study found that psoas muscle volume peaks in the 30s and decreases with age, a pattern more pronounced in men.

 

■ "Extracting More Health Data from Captured Images" 

The significance of this study lies in the ability to obtain additional health information from CT scans already performed for other reasons, such as lung cancer screening, abdominal issues, or trauma.

While these images contain vast data on muscles, fat, and blood vessels, most information remains unanalyzed as physicians focus on specific pathologies. Medical AI is now evolving to read this "hidden data". This approach, known as opportunistic screening, allows for the simultaneous analysis of lung disease, cardiovascular risk, and body composition changes from a single scan, even identifying unknown causes of other conditions like diabetes or cancer. This method is gaining significant attention as it increases the chance of early disease detection without increasing costs or radiation exposure.

 

■ The Next Step for Medical AI: Image-Based Health Indicators 

This shift represents a major trend in the medical AI market. While early AI focused on detecting specific diseases, it is now evolving toward analyzing multiple health indicators simultaneously.

In a rapidly aging society, sarcopenia is a critical indicator linked to falls and chronic disease aggravation, yet it is often under-diagnosed in clinical settings. This study provides an objective standard for evaluation. The research team explained, "This study holds clinical value by presenting a new digital indicator to detect sarcopenia early. We aim to open the era of 'opportunistic screening' using existing abdominal CTs and contribute to a precision medical system centered on prevention by quantifying muscle loss trends that begin in the 30s".