Deep Learning Model Shows Promise in Detecting Blunt Liver Injury via Ultrasound
Researchers utilizing an animal model have evaluated a deep-learning-based tool to automatically detect and grade blunt liver injury using ultrasound images.
Key points
- ResNet50 deep-learning model evaluated for blunt liver injury diagnosis
- Study conducted on an animal model using 54 New Zealand rabbits
- Model achieved notable accuracy for detecting hemoperitoneum and trauma focus
A team of researchers has explored the use of a deep-learning network, specifically the ResNet50 model, to analyze ultrasound scans for blunt liver injuries. Published in Nature Health Sciences, the study involved 54 healthy New Zealand rabbits to establish validated standards for blunt liver trauma. Because evaluating ultrasound images typically relies on subjective human interpretation—especially when visceral bleeding is absent—the investigators sought to test whether artificial intelligence could provide an objective and reliable diagnosis by identifying both indirect signs like hemoperitoneum and direct trauma focus.
According to the findings, the AI model achieved notable performance metrics in evaluating the animal models. The ResNet50 system demonstrated an accuracy of 0.842, a sensitivity of 0.901, and a specificity of 0.712 when determining the presence of hemoperitoneum. For identifying direct trauma focus, the model yielded an accuracy of 0.864, sensitivity of 0.852, and specificity of 0.881, while also successfully discriminating the degree of liver injury with an accuracy of 0.732. Additionally, the researchers found a significant positive correlation between the calculated hemoperitoneum index and the actual weight of the hemoperitoneum.
While these results suggest that deep learning may support the identification of liver trauma beyond simple hemoperitoneum detection, the findings remain limited to an animal model. Experts emphasize that further clinical research involving human subjects is necessary before such technology can be implemented in emergency medical settings. Readers are advised to consult qualified healthcare professionals for medical guidance and personal healthcare decisions.
What to know: A new animal study published in Nature Health Sciences indicates that a ResNet50 deep-learning model may accurately assist in diagnosing and grading blunt liver injuries using ultrasound scans.
Source Nature Health Sciences — nature.com/articles/s41598-026-79868-2 ↗
Read next
🏥Public HealthPreventive Services Task Force Resumes Meetings Under New Leadership
The U.S. Preventive Services Task Force is set to reconvene this week for the first time in over a year following a significant membership overhaul by Health Secretary Ro…
❤️Conditions & CareBlood-Based Genetic Signature May Predict CAR T-Cell Therapy Success in Lymphoma
Researchers have identified a specific RNA signature in blood plasma that may help predict how well lymphoma patients respond to CAR T-cell therapy.
🧠Mind & Mental HealthAddressing Body Image and the Reality of Gluteal Anatomy
A personal reflection on the social and psychological impact of body shape, specifically regarding the gluteal region, and the limitations of consumer solutions for perce…
❤️Conditions & CareMigraine and Smoking Linked to Increased Early Stroke Risk in Women
A large-scale study suggests that people who have given birth and experience migraines while also smoking face a significantly higher risk of stroke before age 60.
🔬ResearchCBD Oil Found Ineffective for Managing Behavioral Issues in Youths with Intellectual Disabilities
A double-blind clinical trial published in JAMA Pediatrics indicates that CBD oil provides no significant behavioral improvement for children and teenagers with intellect…
🏃FitnessExpert Advice on Hydration Gear Sparks Debate Among Recreational Runners
Olympic medalist Josh Kerr recently suggested that runners may not require hydration packs for distances under 12 miles, igniting a broader conversation about gear necess…