Practical computer vision application to detect hip fractures on pelvic X-rays: a bi-institutional study

Artificial intelligence diagnostic accuracy in fracture detection from plain radiographs and comparing it with clinicians: a systematic review and meta-analysis

PURPOSE

Fracture detection is one of the most commonly used and studied aspects of artificial intelligence (AI) in medicine. In this systematic review and meta-analysis, we aimed to summarize available literature and data regarding AI performance in fracture detection on plain radiographs and various factors affecting it.

METHODS

We systematically reviewed studies evaluating AI algorithms in detecting bone fractures in plain radiographs, combined their performance using meta-analysis (a bivariate regression approach), and compared it with that of clinicians. We also analyzed the factors potentially affecting algorithm performance using meta-regression.

RESULTS

Our analysis included 100 studies. In 83 studies with confusion matrices, AI algorithms showed a sensitivity of 91.43% and a specificity of 92.12% (Area under the summary receiver operator curve = 0.968). After adjustment and false discovery rate correction, tibia/fibula (excluding ankle) fractures were associated with higher (7.0%, p=0.004) AI sensitivity, while more recent publications (5.5%, p=0.003) and Xception architecture (6.6%, p<0.001) were associated with higher specificity. Clinicians and AI showed similar specificity in fracture identification, although AI leaned to higher sensitivity (7.6%, p=0.07). Radiologists, on the other hand, were more specific than AI overall and in several subgroups, and more sensitive to hip fractures before FDR correction.

CONCLUSIONS

Currently available AI aids could result in a significant improvement in care where radiologists are not readily available. Moreover, identifying factors affecting algorithm performance could guide AI development teams in their process of optimizing their products.

Artificial intelligence in musculoskeletal imaging: realistic clinical applications in the next decade

This article will provide a perspective review of the most extensively investigated deep learning (DL) applications for musculoskeletal disease detection that have the best potential to translate into routine clinical practice over the next decade. Deep learning methods for detecting fractures, estimating pediatric bone age, calculating bone measurements such as lower extremity alignment and Cobb angle, and grading osteoarthritis on radiographs have been shown to have high diagnostic performance with many of these applications now commercially available for use in clinical practice. Many studies have also documented the feasibility of using DL methods for detecting joint pathology and characterizing bone tumors on magnetic resonance imaging (MRI). However, musculoskeletal disease detection on MRI is difficult as it requires multi-task, multi-class detection of complex abnormalities on multiple image slices with different tissue contrasts. The generalizability of DL methods for musculoskeletal disease detection on MRI is also challenging due to fluctuations in image quality caused by the wide variety of scanners and pulse sequences used in routine MRI protocols. The diagnostic performance of current DL methods for musculoskeletal disease detection must be further evaluated in well-designed prospective studies using large image datasets acquired at different institutions with different imaging parameters and imaging hardware before they can be fully implemented in clinical practice. Future studies must also investigate the true clinical benefits of current DL methods and determine whether they could enhance quality, reduce error rates, improve workflow, and decrease radiologist fatigue and burnout with all of this weighed against the costs.

Proximal femur fracture detection on plain radiography via feature pyramid networks

Hip fractures exceed 250,000 cases annually in the United States, with the worldwide incidence projected to increase by 240-310% by 2050. Hip fractures are predominantly diagnosed by radiologist review of radiographs. In this study, we developed a deep learning model by extending the VarifocalNet Feature Pyramid Network (FPN) for detection and localization of proximal femur fractures from plain radiography with clinically relevant metrics. We used a dataset of 823 hip radiographs of 150 subjects with proximal femur fractures and 362 controls to develop and evaluate the deep learning model. Our model attained 0.94 specificity and 0.95 sensitivity in fracture detection over the diverse imaging dataset. We compared the performance of our model against five benchmark FPN models, demonstrating 6-14% sensitivity and 1-9% accuracy improvement. In addition, we demonstrated that our model outperforms a state-of-the-art transformer model based on DINO network by 17% sensitivity and 5% accuracy, while taking half the time on average to process a radiograph. The developed model can aid radiologists and support on-premise integration with hospital cloud services to enable automatic, opportunistic screening for hip fractures.

Developing a Computer Vision Model to Automate Quantitative Measurement of Hip-Knee-Ankle Angle in Total Hip and Knee Arthroplasty Patients

BACKGROUND

Increasing deformity of the lower extremities, as measured by the hip-knee-ankle angle (HKAA), is associated with poor patient outcomes after total hip and knee arthroplasty (THA, TKA). Automated calculation of HKAA is imperative to reduce the burden on orthopaedic surgeons. We proposed a detection-based deep learning (DL) model to calculate HKAA in THA and TKA patients and assessed the agreement between DL-derived HKAAs and manual measurement.

METHODS

We retrospectively identified 1,379 long-leg radiographs (LLRs) from patients scheduled for THA or TKA within an academic medical center. There were 1,221 LLRs used to develop the model (randomly split into 70% training, 20% validation, and 10% held-out test sets); 158 LLRs were considered "difficult," as the femoral head was difficult to distinguish from surrounding tissue. There were 2 raters who annotated the HKAA of both lower extremities, and inter-rater reliability was calculated to compare the DL-derived HKAAs with manual measurement within the test set.

RESULTS

The DL model achieved a mean average precision of 0.985 on the test set. The average HKAA of the operative leg was 173.05 ± 4.54°; the nonoperative leg was 175.55 ± 3.56°. The inter-rater reliability between manual and DL-derived HKAA measurements on the operative leg and nonoperative leg indicated excellent reliability (intraclass correlation (2,k) = 0.987 [0.96, 0.99], intraclass correlation (2, k) = 0.987 [0.98, 0.99, respectively]). The standard error of measurement for the DL-derived HKAA for the operative and nonoperative legs was 0.515° and 0.403°, respectively.

CONCLUSIONS

A detection-based DL algorithm can calculate the HKAA in LLRs and is comparable to that calculated by manual measurement. The algorithm can detect the bilateral femoral head, knee, and ankle joints with high precision, even in patients where the femoral head is difficult to visualize.

Optimizing the Clinical Direction of Artificial Intelligence With Health Policy: A Narrative Review of the Literature

Artificial intelligence (AI) has the ability to completely transform the healthcare industry by enhancing diagnosis, treatment, and resource allocation. To ensure patient safety and equitable access to healthcare, it also presents ethical and practical issues that need to be carefully addressed. Its integration into healthcare is a crucial topic. To realize its full potential, however, the ethical issues around data privacy, prejudice, and transparency, as well as the practical difficulties posed by workforce adaptability and statutory frameworks, must be addressed. While there is growing knowledge about the advantages of AI in healthcare, there is a significant lack of knowledge about the moral and practical issues that come with its application, particularly in the setting of emergency and critical care. The majority of current research tends to concentrate on the benefits of AI, but thorough studies that investigate the potential disadvantages and ethical issues are scarce. The purpose of our article is to identify and examine the ethical and practical difficulties that arise when implementing AI in emergency medicine and critical care, to provide solutions to these issues, and to give suggestions to healthcare professionals and policymakers. In order to responsibly and successfully integrate AI in these important healthcare domains, policymakers and healthcare professionals must collaborate to create strong regulatory frameworks, safeguard data privacy, remove prejudice, and give healthcare workers the necessary training.

The Feasibility and Performance of Total Hip Replacement Prediction Deep Learning Algorithm with Real World Data

(1) Background: Hip degenerative disorder is a common geriatric disease is the main causes to lead to total hip replacement (THR). The surgical timing of THR is crucial for post-operative recovery. Deep learning (DL) algorithms can be used to detect anomalies in medical images and predict the need for THR. The real world data (RWD) were used to validate the artificial intelligence and DL algorithm in medicine but there was no previous study to prove its function in THR prediction. (2) Methods: We designed a sequential two-stage hip replacement prediction deep learning algorithm to identify the possibility of THR in three months of hip joints by plain pelvic radiography (PXR). We also collected RWD to validate the performance of this algorithm. (3) Results: The RWD totally included 3766 PXRs from 2018 to 2019. The overall accuracy of the algorithm was 0.9633; sensitivity was 0.9450; specificity was 1.000 and the precision was 1.000. The negative predictive value was 0.9009, the false negative rate was 0.0550, and the F1 score was 0.9717. The area under curve was 0.972 with 95% confidence interval from 0.953 to 0.987. (4) Conclusions: In summary, this DL algorithm can provide an accurate and reliable method for detecting hip degeneration and predicting the need for further THR. RWD offered an alternative support of the algorithm and validated its function to save time and cost.

Transfer learning-based ensemble convolutional neural network for accelerated diagnosis of foot fractures

The complex shape of the foot, consisting of 26 bones, variable ligaments, tendons, and muscles leads to misdiagnosis of foot fractures. Despite the introduction of artificial intelligence (AI) to diagnose fractures, the accuracy of foot fracture diagnosis is lower than that of conventional methods. We developed an AI assistant system that assists with consistent diagnosis and helps interns or non-experts improve their diagnosis of foot fractures, and compared the effectiveness of the AI assistance on various groups with different proficiency. Contrast-limited adaptive histogram equalization was used to improve the visibility of original radiographs and data augmentation was applied to prevent overfitting. Preprocessed radiographs were fed to an ensemble model of a transfer learning-based convolutional neural network (CNN) that was developed for foot fracture detection with three models: InceptionResNetV2, MobilenetV1, and ResNet152V2. After training the model, score class activation mapping was applied to visualize the fracture based on the model prediction. The prediction result was evaluated by the receiver operating characteristic (ROC) curve and its area under the curve (AUC), and the F1-Score. Regarding the test set, the ensemble model exhibited better classification ability (F1-Score: 0.837, AUC: 0.95, Accuracy: 86.1%) than other single models that showed an accuracy of 82.4%. With AI assistance for the orthopedic fellow, resident, intern, and student group, the accuracy of each group improved by 3.75%, 7.25%, 6.25%, and 7% respectively and diagnosis time was reduced by 21.9%, 14.7%, 24.4%, and 34.6% respectively.

Diagnostic accuracy and potential covariates of artificial intelligence for diagnosing orthopedic fractures: a systematic literature review and meta-analysis

OBJECTIVES

To systematically quantify the diagnostic accuracy and identify potential covariates affecting the performance of artificial intelligence (AI) in diagnosing orthopedic fractures.

METHODS

PubMed, Embase, Web of Science, and Cochrane Library were systematically searched for studies on AI applications in diagnosing orthopedic fractures from inception to September 29, 2021. Pooled sensitivity and specificity and the area under the receiver operating characteristic curves (AUC) were obtained. This study was registered in the PROSPERO database prior to initiation (CRD 42021254618).

RESULTS

Thirty-nine were eligible for quantitative analysis. The overall pooled AUC, sensitivity, and specificity were 0.96 (95% CI 0.94-0.98), 90% (95% CI 87-92%), and 92% (95% CI 90-94%), respectively. In subgroup analyses, multicenter designed studies yielded higher sensitivity (92% vs. 88%) and specificity (94% vs. 91%) than single-center studies. AI demonstrated higher sensitivity with transfer learning (with vs. without: 92% vs. 87%) or data augmentation (with vs. without: 92% vs. 87%), compared to those without. Utilizing plain X-rays as input images for AI achieved results comparable to CT (AUC 0.96 vs. 0.96). Moreover, AI achieved comparable results to humans (AUC 0.97 vs. 0.97) and better results than non-expert human readers (AUC 0.98 vs. 0.96; sensitivity 95% vs. 88%).

CONCLUSIONS

AI demonstrated high accuracy in diagnosing orthopedic fractures from medical images. Larger-scale studies with higher design quality are needed to validate our findings.

KEY POINTS

• Multicenter study design, application of transfer learning, and data augmentation are closely related to improving the performance of artificial intelligence models in diagnosing orthopedic fractures. • Utilizing plain X-rays as input images for AI to diagnose fractures achieved results comparable to CT (AUC 0.96 vs. 0.96). • AI achieved comparable results to humans (AUC 0.97 vs. 0.97) but was superior to non-expert human readers (AUC 0.98 vs. 0.96, sensitivity 95% vs. 88%) in diagnosing fractures.

Artificial Intelligence in Fracture Detection: A Systematic Review and Meta-Analysis

Background Patients with fractures are a common emergency presentation and may be misdiagnosed at radiologic imaging. An increasing number of studies apply artificial intelligence (AI) techniques to fracture detection as an adjunct to clinician diagnosis. Purpose To perform a systematic review and meta-analysis comparing the diagnostic performance in fracture detection between AI and clinicians in peer-reviewed publications and the gray literature (ie, articles published on preprint repositories). Materials and Methods A search of multiple electronic databases between January 2018 and July 2020 (updated June 2021) was performed that included any primary research studies that developed and/or validated AI for the purposes of fracture detection at any imaging modality and excluded studies that evaluated image segmentation algorithms. Meta-analysis with a hierarchical model to calculate pooled sensitivity and specificity was used. Risk of bias was assessed by using a modified Prediction Model Study Risk of Bias Assessment Tool, or PROBAST, checklist. Results Included for analysis were 42 studies, with 115 contingency tables extracted from 32 studies (55 061 images). Thirty-seven studies identified fractures on radiographs and five studies identified fractures on CT images. For internal validation test sets, the pooled sensitivity was 92% (95% CI: 88, 93) for AI and 91% (95% CI: 85, 95) for clinicians, and the pooled specificity was 91% (95% CI: 88, 93) for AI and 92% (95% CI: 89, 92) for clinicians. For external validation test sets, the pooled sensitivity was 91% (95% CI: 84, 95) for AI and 94% (95% CI: 90, 96) for clinicians, and the pooled specificity was 91% (95% CI: 81, 95) for AI and 94% (95% CI: 91, 95) for clinicians. There were no statistically significant differences between clinician and AI performance. There were 22 of 42 (52%) studies that were judged to have high risk of bias. Meta-regression identified multiple sources of heterogeneity in the data, including risk of bias and fracture type. Conclusion Artificial intelligence (AI) and clinicians had comparable reported diagnostic performance in fracture detection, suggesting that AI technology holds promise as a diagnostic adjunct in future clinical practice. Clinical trial registration no. CRD42020186641 © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Cohen and McInnes in this issue.