Science
Standard-plane acquisition rate among novice physicians improves 9.9%; scanning time also reduced AI standardizes scans using expert experience, with potential to significantly improve access to care

Sonoscape Medical’s artificial intelligence (AI)-based musculoskeletal ultrasound system has demonstrated its technological capabilities after being published in a leading radiology journal.
According to Chinese biotechnology publication BioOn on the 30th, results from a multicenter clinical study of an intelligent AI-assisted ultrasound scanning system for the wrist and hand joints, jointly conducted by West China Hospital of Sichuan University and Sonoscape Medical, were published in the international radiology journal Radiology (IF=17.6). The study also included a special commentary by Professor Levon Nazarian, an international expert in musculoskeletal imaging.
The study found that the AI-assisted system directly improved the standard-plane acquisition rate among novice physicians by 9.9%. The number of images meeting standard specifications increased from 241 to 265 out of 270, while the number of unsuitable, nonstandard images fell substantially to approximately one-sixth of the previous level.
In particular, the system significantly reduced total scanning time among physicians in their first to third years of practice (P<0.001). At the same time, it did not impose an additional burden on experienced physicians, demonstrating its efficiency in clinical practice.
The study was conducted using data from 430 healthy participants at 20 medical centers in China. It used a total of 66,743 static-image datasets to reflect a range of clinical settings and physician experience levels.
In his commentary, Professor Levon Nazarian of the Hospital of the University of Pennsylvania said, "This achievement uses AI to break down the professional barriers to musculoskeletal ultrasound, transforming a technique mastered by only a small number of experts into a standardized, universal medical tool."
Sonoscape has spent 9 years developing musculoskeletal ultrasound AI technology. Beginning with shoulder-joint modeling in 2017, the company developed dynamic ankle-joint tracking technology and has now completed its wrist-joint algorithm. The technology is incorporated into the company’s premium ultrasound system, the HETU (河图) S80 series.
The company is expanding its AI solutions beyond musculoskeletal applications into areas including obstetrics and gynecology, gastrointestinal endoscopy, and cardiovascular interventional procedures. Sonoscape said, "Our goal is to convert the experience of leading experts into replicable intelligent tools and improve access to high-quality medical services."

BIOBOOK TEAM
biobookmedia@biobook.co.kr
Hyundai Bioscience’s Xafty to Enter Phase 2 SFTS Trial Amid Questions Over Patient Recruitment and Efficacy Testing

Google DeepMind Develops AI Protein “Watermark,” Demonstrates Function Is Preserved

BioNTech’s mRNA Cancer Vaccine Fails Phase 3 Endpoint; Shares Fall 5%

First Keytruda Biosimilar Enters FDA Review as Formycon, Zydus File Application

15-Month Follow-Up of In Vivo CAR-T Therapy: 1 of 4 Patients With Multiple Myeloma Maintains Complete Remission

Drug Reverses Damage From APOE4, Alzheimer’s Strongest Genetic Risk Factor, in Experimental Models