Lumbar spine diseases substantially impact the patients'' quality of life, necessitating accurate and efficient diagnostic tools. This study presents Lumbar VNet Pro (LVP), the first real-time artificial-intelligence (AI)-assisted system embedded within MRI hardware for lumbar spine analysis, integrating deep learning with MRI. LVP was trained on 2,453 MRI datasets and validated both internally and externally across multiple centers. During the training (1,848 MRI datasets) and validation (605 MRI datasets), LVP exhibited outstanding performance in localization (Dice = 0.93), segmentation (Dice = 0.92), labeling (identification rate = 0.90), and timeliness (average inference time = 1.1 s). Following the successful construction of LVP, we conducted comprehensive testing through both internal and external multicenter evaluations. Internal testing involving 100 patients indicated that the recognition accuracy of LVP was as high as 100%, and the consistency between the LVP assessment and the manual assessment using the gold standard reached 97%. In external testing involving 1,522 patients, LVP''s diagnostic performance was compared to those of manual and human-machine-assisted methods. The AI-assisted approaches demonstrated better performance across multiple spinal pathologies, including lumbar disc herniation, spinal canal stenosis, and lateral recess stenosis, with area under the receiver operating characteristic curve values >0.95 for deep learning/human-machine approaches and >0.90 for the fully manual approach. The real-time integration of LVP with MRI scanning improved positioning accuracy and reduced interobserver variability, supporting its potential as an adjunct tool for enhancing MRI-based spine diagnostics. However, further studies are warranted to assess its generalizability across diverse clinical settings.
[1]Southern Med Univ, Guangdong Prov Peoples Hosp, Guangdong Acad Med Sci, Dept Spine Surg, Guangzhou 510080, Guangdong, Peoples R China.
[2]Southern Med Univ, Zhujiang Hosp, Dept Radiol, Guangzhou 510280, Peoples R China.
[3]Rhein Westfal TH Aachen, Fac Nat Sci, D-52074 Aachen, Germany.
[4]Hosp Nacl Paraplej, Lab Regenerac Neuronal, IDISCAM, Toledo 45071, Spain.
[5]Sun Yat Sen Univ, Affiliated Hosp 1, Dept Spine Surg, Guangzhou 510000, Peoples R China.
[6]Nanjing Univ, Nanjing Drum Tower Hosp, Affiliated Hosp, Dept Radiol,Med Sch, Nanjing 210008, Peoples R China.
[7]Shanghai United Imaging Healthcare Co Ltd, Healthcare Adv Algorithm Dept HSW BU, Shanghai 201800, Peoples R China.
[8]Shanghai Jiao Tong Univ, Sch Elect Informat & Elect Engn, Shanghai 200240, Peoples R China.
[9]Hannover Med Sch MHH, Inst Neuroanat & Cell Biol, D-30625 Hannover, Germany.
[10]Natl Univ Singapore, Dept Diagnost Radiol, Singapore 119077, Singapore.
[11]United Christian Hosp, Dept Orthopaed & Traumatol, Hong Kong 999077, Peoples R China.
[12]Southern Med Univ, Shunde Hosp, Shunde Peoples Hosp 1, Dept Radiol, Foshan 528300, Peoples R China.
[13]Fudan Univ, Sch Informat Sci & Technol, Shanghai 200438, Peoples R China.
[14]Qiqihar Med Univ, Affiliated Hosp 1, Dept Radiol, Qiqihar 161041, Peoples R China.
[15]Southern Med Univ, Clin Med Coll 2, Guangzhou 510515, Peoples R China.
[16]Southern Med Univ, Nanfang Hosp, Dept Spine Surg, Guangzhou 510515, Peoples R China.
[17]Zhejiang Univ, Affiliated Hosp 1, Sch Med, Dept Orthoped, Hangzhou 310003, Peoples R China.
[18]Tongji Univ, Shanghai Peoples Hosp 10, Sch Med, Dept Radiol, Shanghai, Peoples R China.
(8.0+极速模式)