Artificial intelligence in orthopedic surgery: evolution, current state and future directions

Kurmis, Andrew P.; Ianunzio, Jamie R

doi:10.1186/s42836-022-00112-z

Cited by 58 publications

(49 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Historical two dimensional (2D) templating and planning has already been shown to be far less accurate than modern 3D equivalents [119][120][121]. The evolution to more universal 3D standards is likely to incrementally improve surgical planning [122][123][124] as such technologies become more mainstream. The cutting edge integration of artificial intelligence algorithms into the pre-operative decision making pathways may represent further advancement still [124].…”

Section: Looking To the Future …mentioning

confidence: 99%

“…The evolution to more universal 3D standards is likely to incrementally improve surgical planning [122][123][124] as such technologies become more mainstream. The cutting edge integration of artificial intelligence algorithms into the pre-operative decision making pathways may represent further advancement still [124]. Similarly, as realtime computer navigation is taken up more broadly many anticipate improved attainment of target cup placement [87] (in a similar fashion to accuracy improvements that were seen during the evolution of TKA navigation).…”

Section: Looking To the Future …mentioning

confidence: 99%

See 1 more Smart Citation

Advanced, Imageless Navigation in Contemporary THA: Optimising Acetabular Component Placement

Kurmis¹

2023

Arthroplasty - Advanced Techniques and Future Perspectives

View full text Add to dashboard Cite

Total hip arthroplasty (THA) stands as a reliable and effective way to manage end-stage hip disease secondary to a number of aetiologic conditions. While target ‘safe zones’ are widely quoted and endorsed, an increasingly robust body of evidence suggests that such idealised implantation goals have limited utility in patient-to-patient considerations and that even with a precise goal in mind, surgeons perform inconsistently in achieving these targets intra-operatively. Inter-patient variability, the concept of ‘functional’ safe zones and the largely under-appreciated impact of poor patient positioning (and progressive loss of position during the case) are all recognised and evidence-supported opponents of conventional ‘40/15’ approaches. In an environment whereby accountable cost utility, maximised surgical consistency (i.e., outlier minimisation), improved attainment of target position, and awareness of the radiation exposure burden of many pre-operative templating regimes are all paramount, there appears to be an increasing role for the application of imageless ‘mini’ intra-operative navigation systems for primary (and revision) THA procedures. This chapter reviews the evolution of THA navigation and discusses contemporary applications, defines the challenges associated with unanticipated pelvic movement, and explores potential future directions in the use of this exciting technology.

show abstract

Section: Looking To the Future …mentioning

confidence: 99%

Section: Looking To the Future …mentioning

confidence: 99%

Advanced, Imageless Navigation in Contemporary THA: Optimising Acetabular Component Placement

Kurmis¹

2023

Arthroplasty - Advanced Techniques and Future Perspectives

View full text Add to dashboard Cite

show abstract

“…The models can identify images via digital processing techniques to make the artificial intelligence process more accurate and cost-effective [ 13 ]. The technology includes knee joint recognition and image processing based on deep learning [ 14 ]. Swiecicki et al [ 15 ] developed a diagnostic model based on the two-stage Faster R-CNN model to assess the severity of KOA from both posterior–anterior (PA) and lateral (LAT) views.…”

Section: Introductionmentioning

confidence: 99%

Automatic assessment of knee osteoarthritis severity in portable devices based on deep learning

Yang

et al. 2022

J Orthop Surg Res

View full text Add to dashboard Cite

Background For knee osteoarthritis, the commonly used radiology severity criteria Kellgren–Lawrence lead to variability among surgeons. Most existing diagnosis models require preprocessed radiographs and specific equipment. Methods All enrolled patients diagnosed with KOA who met the criteria were obtained from **** Hospital. This study included 2579 images shot from posterior–anterior X-rays of 2,378 patients. We used RefineDet to train and validate this deep learning-based diagnostic model. After developing the model, 823 images of 697 patients were enrolled as the test set. The whole test set was assessed by up to 5 surgeons and this diagnostic model. To evaluate the model’s performance we compared the results of the model with the KOA severity diagnoses of surgeons based on K-L scales. Results Compared to the diagnoses of surgeons, the model achieved an overall accuracy of 0.977. Its sensitivity (recall) for K-L 0 to 4 was 1.0, 0.972, 0.979, 0.983 and 0.989, respectively; for these diagnoses, the specificity of this model was 0.992, 0.997, 0.994, 0.991 and 0.995. The precision and F1-score were 0.5 and 0.667 for K-L 0, 0.914 and 0.930 for K-L 1, 0.978 and 0.971 for K-L 2, 0.981 and 0.974 for K-L 3, and 0.988 and 0.985 for K-L 4, respectively. All K-L scales perform AUC > 0.90. The quadratic weighted Kappa coefficient between the diagnostic model and surgeons was 0.815 (P < 0.01, 95% CI 0.727–0.903). The performance of the model is comparable to the clinical diagnosis of KOA. This model improved the efficiency and avoided cumbersome image preprocessing. Conclusion The deep learning-based diagnostic model can be used to assess the severity of KOA in portable devices according to the Kellgren–Lawrence scale. On the premise of improving diagnostic efficiency, the results are highly reliable and reproducible.

show abstract

“…The models can identify images via digital processing techniques to make the arti cial intelligence process more accurate and cost-effective 13 . The technology includes knee joint recognition and image processing based on deep learning 14 . Swiecicki et al 15 have developed an diagnostic model based on the 2 stage model Faster R-CNN to assess the severity of KOA from both posterior-anterior (PA) and lateral (LAT) views.…”

Section: Introductionmentioning

confidence: 99%

“…In clinical use, different doctors or the same doctor at different times may often get similar results rather than identical ones on the same X-ray. Many clinical studies involving KOA diagnosis have ensured the reliability by increasing the number of repeated diagnoses [12][13][14][15][16] . Moreover, the total numbers of X-ray examinations are much high in large hospitals, which is a heavy burden on radiologists and surgeons.…”

Section: Introductionmentioning

confidence: 99%

Automatic Assessment of Knee Osteoarthritis Severity in Portable Devices based on Deep Learning

Yang

et al. 2022

Preprint

View full text Add to dashboard Cite

Background For knee osteoarthritis, the commonly used radiology severity criteria Kellgren-Lawrence lead to variability among surgeons. And most existing diagnosis models require pre-processed radiographs and specific equipment. Methods All enrolled patients diagnosed as KOA who met the criteria were obtained from **** hospital. This study included 2579 images shot from posterior-anterior X-rays of 2378 patients. We used the RefineDet to train and validate this deep learning-based diagnostic model. After developing it, 823 images of 697 patients were enrolled as test set. The whole test set were assessed by up to 5 surgeons and this diagnostic model respectively. To evaluate the model’s performance. we compared the results of model with the KOA severity diagnoses of surgeons based on K-L scales. Results Compared to diagnoses of surgeons, the model achieved an overall accuracy of 0.977. Its sensitivity(recall) for K-L 0 to 4 was 1.0, 0.972, 0.979, 0.983 and 0.989 respectively; for these diagnoses, the specificity of this model was 0.992, 0.997, 0.994, 0.991 and 0.995. The precision and F1-score were 0.5 and 0.667 for K-L 0, 0.914 and 0.930 for K-L 1, 0.978 and 0.971 for K-L 2, 0.981 and 0.974 for K-L 3, 0.988 and 0.985 for K-L 4. And all K-L scales perform AUC > 0.90.The quadratic weighted Kappa coefficient between the diagnostic model and surgeons was 0.815(P＜0.01, 95%CI 0.727–0.903). The performance of the model is comparable to clinical diagnosis of KOA. This model improved the effciency and avoid the cumbersome image-preprocessing. Conclusion The deep learning-based diagnostic model can be used to assess the severity of knee osteoarthritis in portable devices according to Kellgren-Lawrence scales. On the premise of improving the diagnostic efficiency, The results are high-reliability and reproducibility.

show abstract

Artificial intelligence in orthopedic surgery: evolution, current state and future directions

Cited by 58 publications

References 43 publications

Advanced, Imageless Navigation in Contemporary THA: Optimising Acetabular Component Placement

Advanced, Imageless Navigation in Contemporary THA: Optimising Acetabular Component Placement

Automatic assessment of knee osteoarthritis severity in portable devices based on deep learning

Automatic Assessment of Knee Osteoarthritis Severity in Portable Devices based on Deep Learning

Contact Info

Product

Resources

About