Lu Yining, Yang Linjun, Mulford Kellen, Grove Austin, Kaji Ellie, Pareek Ayoosh, Levy Bruce, Wyles Cody C, Camp Christopher L, Krych Aaron J
Department of Orthopedic Surgery, Mayo Clinic, Rochester, Minnesota, USA.
Orthopedic Surgery Artificial Intelligence Laboratory, Mayo Clinic, Rochester, Minnesota, USA.
Knee Surg Sports Traumatol Arthrosc. 2025 Feb 10. doi: 10.1002/ksa.12618.
Developing large-scale, standardized radiographic registries for anterior cruciate ligament (ACL) injuries with artificial intelligence (AI) tools can enhance personalized orthopaedics. We propose deploying Artificial Intelligence for Knee Imaging Registration and Analysis (AKIRA), a trio of deep learning (DL) algorithms, to automatically classify and annotate radiographs. We hypothesize that algorithms can efficiently organize radiographs based on laterality, projection, identify implants and classify osteoarthritis (OA) grade.
A collection of 20,836 knee radiographs from all time points of treatment (mean orthopaedic follow-up 70.7 months; interquartile range [IQR]: 6.8-172 months) were aggregated from 1628 ACL-injured patients (median age 26 years [IQR: 19-42], 57% male). Three DL algorithms (EfficientNet, YOLO [You Only Look Once] and Residual Network) were employed. Radiograph laterality and projection (anterior-posterior [AP], lateral, sunrise, posterior-anterior, hip-knee-ankle and Camp-Coventry intercondylar [notch]) were labelled by a DL model. Manually provided labels of metal fixation implants were used to develop a DL object detection algorithm. The degree of OA, both as measured by specific Kellgren-Lawrence (KL) grades, as well as based on a binarized label of OA (defined as KL Grade ≥2), on standing AP radiographs were classified using a DL algorithm. Individual model performances were evaluated on a subset of images prior to the deployment of AKIRA to registry construction using all ACL radiographs.
The classification algorithms showed excellent performance in classifying radiographic laterality (F1 score: 0.962-0.975) and projection (F1 score: 0.941-1.0). The object detection algorithm achieved high precision-recall (area under the precision-recall curve: 0.695-0.992) for identifying various metal fixations. The KL classifier reached concordances of 0.39-0.40, improving to 0.81-0.82 for binary OA labels. Sequential deployment of AKIRA following internal validation processed and labelled all 20,836 images with the appropriate views, implants, and the presence of OA within 88 min.
AKIRA effectively automated the classification and object detection in a large radiograph cohort of ACL injuries, creating an AI-enabled radiographic registry with comprehensive details on laterality, projection, implants and OA.
Cross-sectional study.
Level IV.
利用人工智能（AI）工具开发大规模、标准化的前交叉韧带（ACL）损伤影像学注册库，可促进个性化骨科治疗。我们提议部署用于膝关节成像配准和分析的人工智能（AKIRA），这是一组深度学习（DL）算法，用于自动对X线片进行分类和标注。我们假设这些算法能够根据左右侧、投照方式有效整理X线片，识别植入物并对骨关节炎（OA）分级。
从1628例ACL损伤患者（年龄中位数26岁[四分位间距：19 - 42岁]，57%为男性）中汇总收集了20836张治疗各时间点的膝关节X线片（骨科平均随访70.7个月；四分位间距[IQR]：6.8 - 172个月）。采用了三种DL算法（EfficientNet、YOLO[你只看一次]和残差网络）。通过一个DL模型对X线片的左右侧和投照方式（前后位[AP]、侧位、日出位、后前位、髋膝踝位和坎普 - 考文垂髁间[切迹]位）进行标注。使用人工提供的金属固定植入物标签来开发一个DL目标检测算法。使用一个DL算法对站立位AP X线片上根据特定的凯尔格伦 - 劳伦斯（KL）分级以及基于OA的二分类标签（定义为KL分级≥2）所测量的OA程度进行分类。在将AKIRA部署到使用所有ACL X线片构建注册库之前，在图像子集中评估各个模型的性能。
分类算法在对X线片的左右侧进行分类时表现出色（F1分数：0.962 - 0.975），在投照方式分类方面也表现出色（F1分数：0.941 - 1.0）。目标检测算法在识别各种金属固定物方面实现了高精度召回率（精确召回率曲线下面积：0.695 - 0.992）。KL分级器的一致性为0.39 - 0.40，对于二分类OA标签，一致性提高到0.81 - 0.82。经过内部验证后依次部署AKIRA，在88分钟内对所有20836张图像进行了适当视图、植入物以及OA存在情况的处理和标注。
AKIRA有效地实现了对一大组ACL损伤X线片的分类和目标检测自动化，创建了一个具有左右侧、投照方式、植入物和OA详细综合信息的人工智能辅助影像学注册库。
横断面研究。
IV级。
