{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,9]],"date-time":"2025-12-09T06:38:32Z","timestamp":1765262312471,"version":"3.46.0"},"reference-count":21,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,12,9]],"date-time":"2025-12-09T00:00:00Z","timestamp":1765238400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>Point of care ultrasound (POCUS) is commonly used for diagnostic triage of internal injuries in both civilian and military trauma. In resource constrained environments, such as mass-casualty situations on the battlefield, POCUS allows medical providers to rapidly and noninvasively assess for free fluid or hemorrhage induced by trauma. A major disadvantage of POCUS diagnostics is the skill threshold needed to acquire and interpret ultrasound scans. For this purpose, AI has been shown to be an effective tool to aid the caregiver when interpreting medical imaging. Here, we focus on sophisticated AI training methodologies to improve the blind, real-time diagnostic accuracy of AI models for detection of hemorrhage in two major abdominal scan sites. In this work, we used a retrospective dataset of over 60,000 swine ultrasound images to train binary classification models exploring frame-pooling methods using the backbone of a pre-existing model architecture to handle multi-channel inputs for detecting free fluid in the pelvic and right-upper-quadrant regions. Earlier classifications models had achieved 0.59 and 0.70 accuracy metrics in blind predictions, respectively. After implementing this novel training technique, performance accuracy improved to over 0.90 for both scan sites. These are promising results demonstrating a significant diagnostic improvement which encourages further optimization to achieve similar results using clinical data. Furthermore, these results show how AI-informed diagnostics can offload cognitive burden in situations where casualties may benefit from rapid triage decision making.<\/jats:p>","DOI":"10.3389\/frai.2025.1718503","type":"journal-article","created":{"date-parts":[[2025,12,9]],"date-time":"2025-12-09T06:29:00Z","timestamp":1765261740000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Adaptation of convolutional neural networks for real-time abdominal ultrasound interpretation"],"prefix":"10.3389","volume":"8","author":[{"given":"Austin J.","family":"Ruiz","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sof\u00eda I.","family":"Hern\u00e1ndez Torres","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Eric J.","family":"Snider","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1965","published-online":{"date-parts":[[2025,12,9]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","first-page":"1775","DOI":"10.1007\/s00261-024-04640-x","article-title":"Artificial intelligence in abdominal and pelvic ultrasound imaging: current applications","volume":"50","author":"Cai","year":"2025","journal-title":"Abdom Radiol"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"707437","DOI":"10.3389\/fmed.2021.707437","article-title":"Deep learning assisted detection of abdominal free fluid in Morison\u2019s pouch during focused assessment with sonography in trauma","volume":"8","author":"Cheng","year":"2021","journal-title":"Front Med"},{"key":"ref3","doi-asserted-by":"publisher","first-page":"S124","DOI":"10.1097\/TA.0000000000003308","article-title":"Point-of-care ultrasound for treatment and triage in austere military environments","volume":"91","author":"Dubecq","year":"2021","journal-title":"J. Trauma Acute Care Surg."},{"key":"ref4","doi-asserted-by":"publisher","first-page":"374","DOI":"10.1053\/j.sult.2018.03.007","article-title":"Point-of-care ultrasound in trauma","volume":"39","author":"Gleeson","year":"2018","journal-title":"Semin. Ultrasound CT MRI"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"29","DOI":"10.3390\/technologies13010029","article-title":"Real-time deployment of ultrasound image interpretation AI models for emergency medicine triage using a swine model","volume":"13","author":"Hernandez Torres","year":"2025","journal-title":"Technologies"},{"key":"ref6","doi-asserted-by":"publisher","first-page":"392","DOI":"10.3390\/bioengineering11040392","article-title":"Evaluation of deep learning model architectures for point-of-care ultrasound diagnostics","volume":"11","author":"Hernandez Torres","year":"2024","journal-title":"Bioengineering"},{"key":"ref7","article-title":"MobileNets: efficient convolutional neural networks for mobile vision applications","author":"Howard","year":"2017"},{"key":"ref8","doi-asserted-by":"crossref","DOI":"10.1109\/ICCV.2019.00140","article-title":"Searching for MobileNetV3","author":"Howard","year":"2019"},{"key":"ref9","article-title":"Adam: a method for stochastic optimization","author":"Kingma","year":"2017"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"1915","DOI":"10.1002\/jum.15868","article-title":"Development and validation of a deep learning strategy for automated view classification of pediatric focused assessment with sonography for trauma","volume":"41","author":"Kornblith","year":"2022","journal-title":"J. Ultrasound Med."},{"key":"ref11","doi-asserted-by":"publisher","first-page":"2035","DOI":"10.1007\/s10278-023-00845-6","article-title":"Using deep learning to detect the presence and location of Hemoperitoneum on the focused assessment with sonography in trauma (FAST) examination in adults","volume":"36","author":"Leo","year":"2023","journal-title":"J. Digit. Imaging"},{"key":"ref12","doi-asserted-by":"publisher","first-page":"e13695","DOI":"10.1002\/acm2.13695","article-title":"Deep learning for emergency ascites diagnosis using ultrasonography images","volume":"23","author":"Lin","year":"2022","journal-title":"J. Appl. Clin. Med. Phys."},{"key":"ref13","doi-asserted-by":"publisher","first-page":"270","DOI":"10.5811\/westjem.2016.11.30435","article-title":"Caudal edge of the liver in the right upper quadrant (RUQ) view is the Most sensitive area for free fluid on the FAST exam","volume":"18","author":"Lobo","year":"2017","journal-title":"West. J. Emerg. Med."},{"key":"ref14","volume-title":"GitHub","year":""},{"key":"ref15","doi-asserted-by":"publisher","first-page":"222","DOI":"10.3390\/jimaging11070222","article-title":"Development of deep learning models for real-time thoracic ultrasound image interpretation","volume":"11","author":"Ruiz","year":"2025","journal-title":"J. Imaging"},{"key":"ref16","doi-asserted-by":"publisher","first-page":"21","DOI":"10.7205\/MILMED-D-12-00267","article-title":"Ultrasound in the austere environment: a review of the history, indications, and specifications","volume":"178","author":"Russell","year":"2013","journal-title":"Mil. Med."},{"key":"ref17","doi-asserted-by":"publisher","first-page":"466","DOI":"10.1097\/00005373-199903000-00022","article-title":"Focused assessment with sonography for trauma (FAST): results from an international consensus conference","volume":"46","author":"Scalea","year":"1999","journal-title":"J. Trauma"},{"key":"ref18","article-title":"Recurrent neural networks (RNNs): a gentle introduction and overview","author":"Schmidt","year":"2019"},{"key":"ref19","article-title":"Efficientnet: rethinking model scaling for convolutional neural networks","author":"Tan","year":"2020"},{"key":"ref20","doi-asserted-by":"publisher","first-page":"e0310404","DOI":"10.1371\/journal.pone.0310404","article-title":"A qualitative study of perceived barriers and facilitators to point-of-care ultrasound use among veterans affairs emergency department providers","volume":"19","author":"Theophanous","year":"2024","journal-title":"PLoS One"},{"key":"ref21","volume-title":"The U.S. Army in Multi-Domain Operations 2028","author":"Townsend","year":"2018"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1718503\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,9]],"date-time":"2025-12-09T06:29:02Z","timestamp":1765261742000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1718503\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,9]]},"references-count":21,"alternative-id":["10.3389\/frai.2025.1718503"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1718503","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,9]]},"article-number":"1718503"}}