%0 Journal Article
%T Research on the Diagnostic Efficacy of AI Software Combined with Thin-Slice CT in Occult Toe Fractures
%A Dazhi Nong
%A Kangying Pang
%A Lijun Luo
%A Shiwen Luo
%A Xinjuan Deng
%A Yuhua Guo
%J Journal of Biomedical Science and Engineering
%P 98-105
%@ 1937-688X
%D 2025
%I Scientific Research Publishing
%R 10.4236/jbise.2025.184007
%X Objective: To evaluate the efficacy of AI software combined with thin-slice CT in the diagnosis of occult toe fractures. Methods: A retrospective analysis was conducted on imaging data from 104 patients with suspected toe fractures at the Third People’s Hospital of Nanning from January 2017 to December 2024. Comparisons were made between CT manual diagnosis and AI-assisted diagnosis. Statistical analysis was performed using SPSS 27.0 to compare the detection rates and diagnostic efficacy of the two methods. Results: The detection rates of AI-assisted diagnosis were significantly higher than those of CT manual diagnosis across all fracture sites. In metatarsal fractures, AI-assisted diagnosis detected 30 cases in the first metatarsal, 25 in the second, 22 in the third, 20 in the fourth, and 18 in the fifth, all higher than the 25, 20, 18, 15, and 12 cases detected by CT manual diagnosis (P < 0.05). For phalangeal fractures, AI-assisted diagnosis showed significant improvements in detection rates, particularly in the proximal phalanx of the first toe (28 vs. 22 cases) and the distal phalanx of the first toe (24 vs. 18 cases), with more occult micro-fractures detected in the proximal phalanx of the fifth toe. In cuneiform fractures, AI-assisted diagnosis detected 15 cases in the medial cuneiform, 12 in the intermediate, and 18 in the lateral, compared to 10, 8, and 12 cases by CT manual diagnosis (P < 0.05). Additionally, AI-assisted diagnosis showed higher detection rates in the navicular (10 vs. 6 cases), cuboid (12 vs. 8 cases), calcaneus (6 vs. 4 cases), and soft tissue injuries (15 vs. 10 cases) (P < 0.05). Overall, the total detection rate of AI-assisted diagnosis was 392 cases, significantly higher than the 278 cases detected by CT manual diagnosis (P < 0.0001). Conclusion: AI software combined with thin-slice CT demonstrates significant advantages in diagnosing occult toe fractures, improving detection rates and providing more reliable clinical diagnostic evidence. However, AI diagnostic results should be integrated with clinical judgment, and further optimization of algorithms and data is needed to enhance accuracy and applicability.
%K AI Software
%K Thin-Slice CT
%K Occult Toe Fractures
%K Diagnostic Efficacy
%U http://www.scirp.org/journal/PaperInformation.aspx?PaperID=142105