Opportunistic osteoporosis screening in multi-detector CT images using deep convolutional neural networks

Effect of fully automatic classification model from different tube voltage images on bone density screening: A self-controlled study

PURPOSE

To develop two bone status prediction models combining deep learning and radiomics based on standard-dose chest computed tomography (SDCT) and low-dose chest computed tomography (LDCT), and to evaluate the effect of tube voltage on reproducibility of radiomics features and predictive efficacy of these models.

METHODS

A total of 1508 patients were enrolled in this retrospective study. LDCT was conducted using 80 kVp, tube current ranging from 100 to 475 mA. On the other hand, SDCT was performed using 120 kVp, tube current ranging from 100 to 520 mA. We developed an automatic thoracic vertebral cancellous bone (TVCB) segmentation model. Subsequently, 1184 features were extracted and two classifiers were developed based on LDCT and SDCT images. Based on the diagnostic results of quantitative computed tomography examination, the first-level classifier was initially developed to distinguish normal or abnormal BMD (including osteoporosis and osteopenia), while the second-level classifier was employed to identify osteoporosis or osteopenia. The Dice coefficient was used to evaluate the performance of the automated segmentation model. The Concordance Correlation Coefficients (CCC) of radiomics features were calculated between LDCT and SDCT, and the performance of these models was evaluated.

RESULTS

Our automated segmentation model achieved a Dice coefficient of 0.98 ± 0.01 and 0.97 ± 0.02 in LDCT and SDCT, respectively. Alterations in tube voltage decreased the reproducibility of the extracted radiomic features, with 85.05 % of the radiomic features exhibiting low reproducibility (CCC < 0.75). The area under the curve (AUC) using LDCT-based and SDCT-based models was 0.97 ± 0.01 and 0.94 ± 0.02, respectively. Nonetheless, cross-validation with independent test sets of different tube voltage scans suggests that variations in tube voltage can impair the diagnostic efficacy of the model. Consequently, radiomics models are not universally applicable to images of varying tube voltages. In clinical settings, ensuring consistency between the tube voltage of the image used for model development and that of the acquired patient image is critical.

CONCLUSIONS

Automatic bone status prediction models, utilizing either LDCT or SDCT images, enable accurate assessment of bone status. Tube voltage impacts reproducibility of features and predictive efficacy of models. It is necessary to account for tube voltage variation during the image acquisition.

Deep learning in the radiologic diagnosis of osteoporosis: a literature review

OBJECTIVE

Osteoporosis is a systemic bone disease characterized by low bone mass, damaged bone microstructure, increased bone fragility, and susceptibility to fractures. With the rapid development of artificial intelligence, a series of studies have reported deep learning applications in the screening and diagnosis of osteoporosis. The aim of this review was to summary the application of deep learning methods in the radiologic diagnosis of osteoporosis.

METHODS

We conducted a two-step literature search using the PubMed and Web of Science databases. In this review, we focused on routine radiologic methods, such as X-ray, computed tomography, and magnetic resonance imaging, used to opportunistically screen for osteoporosis.

RESULTS

A total of 40 studies were included in this review. These studies were divided into three categories: osteoporosis screening (n = 20), bone mineral density prediction (n = 13), and osteoporotic fracture risk prediction and detection (n = 7).

CONCLUSIONS

Deep learning has demonstrated a remarkable capacity for osteoporosis screening. However, clinical commercialization of a diagnostic model for osteoporosis remains a challenge.

Cost-Effectiveness of Artificial Intelligence-Based Opportunistic Compression Fracture Screening of Existing Radiographs

PURPOSE

Osteoporotic vertebral compression fractures (OVCFs) are a highly prevalent source of morbidity and mortality, and preventive treatment has been demonstrated to be both effective and cost effective. To take advantage of the information available on existing chest and abdominal radiographs, the authors' study group has developed software to access these radiographs for OVCFs with high sensitivity and specificity using an established artificial intelligence deep learning algorithm. The aim of this analysis was to assess the potential cost-effectiveness of implementing this software.

METHODS

A deterministic expected-value cost-utility model was created, combining a tree model and a Markov model, to compare the strategies of opportunistic screening for OVCFs against usual care. Total costs and total quality-adjusted life-years were calculated for each strategy. Screening and treatment costs were considered from a limited societal perspective, at 2022 prices.

RESULTS

In the base case, assuming a cost of software implantation of $10 per patient screened, the screening strategy dominated the nonscreening strategy: it resulted in lower cost and increased quality-adjusted life-years. The lower cost was due primarily to the decreased costs associated with fracture treatment and decreased probability of requiring long-term care in patients who received preventive treatment. The screening strategy was dominant up to a cost of $46 per patient screened.

CONCLUSIONS

Artificial intelligence-based opportunistic screening for OVCFs on existing radiographs can be cost effective from a societal perspective.

Hybrid transformer convolutional neural network-based radiomics models for osteoporosis screening in routine CT

OBJECTIVE

Early diagnosis of osteoporosis is crucial to prevent osteoporotic vertebral fracture and complications of spine surgery. We aimed to conduct a hybrid transformer convolutional neural network (HTCNN)-based radiomics model for osteoporosis screening in routine CT.

METHODS

To investigate the HTCNN algorithm for vertebrae and trabecular segmentation, 92 training subjects and 45 test subjects were employed. Furthermore, we included 283 vertebral bodies and randomly divided them into the training cohort (n = 204) and test cohort (n = 79) for radiomics analysis. Area receiver operating characteristic curves (AUCs) and decision curve analysis (DCA) were applied to compare the performance and clinical value between radiomics models and Hounsfield Unit (HU) values to detect dual-energy X-ray absorptiometry (DXA) based osteoporosis.

RESULTS

HTCNN algorithm revealed high precision for the segmentation of the vertebral body and trabecular compartment. In test sets, the mean dice scores reach 0.968 and 0.961. 12 features from the trabecular compartment and 15 features from the entire vertebral body were used to calculate the radiomics score (rad score). Compared with HU values and trabecular rad-score, the vertebrae rad-score suggested the best efficacy for osteoporosis and non-osteoporosis discrimination (training group: AUC = 0.95, 95%CI 0.91-0.99; test group: AUC = 0.97, 95%CI 0.93-1.00) and the differences were significant in test group according to the DeLong test (p < 0.05).

CONCLUSIONS

This retrospective study demonstrated the superiority of the HTCNN-based vertebrae radiomics model for osteoporosis discrimination in routine CT.

DeepmdQCT: A multitask network with domain invariant features and comprehensive attention mechanism for quantitative computer tomography diagnosis of osteoporosis

In the medical field, the application of machine learning technology in the automatic diagnosis and monitoring of osteoporosis often faces challenges related to domain adaptation in drug therapy research. The existing neural networks used for the diagnosis of osteoporosis may experience a decrease in model performance when applied to new data domains due to changes in radiation dose and equipment. To address this issue, in this study, we propose a new method for multi domain diagnostic and quantitative computed tomography (QCT) images, called DeepmdQCT. This method adopts a domain invariant feature strategy and integrates a comprehensive attention mechanism to guide the fusion of global and local features, effectively improving the diagnostic performance of multi domain CT images. We conducted experimental evaluations on a self-created OQCT dataset, and the results showed that for dose domain images, the average accuracy reached 91%, while for device domain images, the accuracy reached 90.5%. our method successfully estimated bone density values, with a fit of 0.95 to the gold standard. Our method not only achieved high accuracy in CT images in the dose and equipment fields, but also successfully estimated key bone density values, which is crucial for evaluating the effectiveness of osteoporosis drug treatment. In addition, we validated the effectiveness of our architecture in feature extraction using three publicly available datasets. We also encourage the application of the DeepmdQCT method to a wider range of medical image analysis fields to improve the performance of multi-domain images.

Effectiveness of opportunistic osteoporosis screening on chest CT using the DCNN model

OBJECTIVE

To develop and evaluate a deep learning model based on chest CT that achieves favorable performance on opportunistic osteoporosis screening using the lumbar 1 + lumbar 2 vertebral bodies fusion feature images, and explore the feasibility and effectiveness of the model based on the lumbar 1 vertebral body alone.

MATERIALS AND METHODS

The chest CT images of 1048 health check subjects from January 2021 to June were retrospectively collected as the internal dataset (the segmentation model: 548 for training, 100 for tuning and 400 for test. The classification model: 530 for training, 100 for validation and 418 for test set). The subjects were divided into three categories according to the quantitative CT measurements, namely, normal, osteopenia and osteoporosis. First, a deep learning-based segmentation model was constructed, and the dice similarity coefficient(DSC) was used to compare the consistency between the model and manual labelling. Then, two classification models were established, namely, (i) model 1 (fusion feature construction of lumbar vertebral bodies 1 and 2) and (ii) model 2 (feature construction of lumbar 1 alone). Receiver operating characteristic curves were used to evaluate the diagnostic efficacy of the models, and the Delong test was used to compare the areas under the curve.

RESULTS

When the number of images in the training set was 300, the DSC value was 0.951 ± 0.030 in the test set. The results showed that the model 1 diagnosing normal, osteopenia and osteoporosis achieved an AUC of 0.990, 0.952 and 0.980; the model 2 diagnosing normal, osteopenia and osteoporosis achieved an AUC of 0.983, 0.940 and 0.978. The Delong test showed that there was no significant difference in area under the curve (AUC) values between the osteopenia group and osteoporosis group (P = 0.210, 0.546), while the AUC value of normal model 2 was higher than that of model 1 (0.990 vs. 0.983, P = 0.033).

CONCLUSION

This study proposed a chest CT deep learning model that achieves favorable performance on opportunistic osteoporosis screening using the lumbar 1 + lumbar 2 vertebral bodies fusion feature images. We further constructed the comparable model based on the lumbar 1 vertebra alone which can shorten the scan length, reduce the radiation dose received by patients, and reduce the training cost of technologists.

Evaluation of deep learning-based quantitative computed tomography for opportunistic osteoporosis screening

To evaluate diagnostic efficacy of deep learning (DL)-based automated bone mineral density (BMD) measurement for opportunistic screening of osteoporosis with routine computed tomography (CT) scans. A DL-based automated quantitative computed tomography (DL-QCT) solution was evaluated with 112 routine clinical CT scans from 84 patients who underwent either chest (N:39), lumbar spine (N:34), or abdominal CT (N:39) scan. The automated BMD measurements (DL-BMD) on L1 and L2 vertebral bodies from DL-QCT were validated with manual BMD (m-BMD) measurement from conventional asynchronous QCT using Pearson's correlation and intraclass correlation. Receiver operating characteristic curve (ROC) analysis identified the diagnostic ability of DL-BMD for low BMD and osteoporosis, determined by dual-energy X-ray absorptiometry (DXA) and m-BMD. Excellent concordance were seen between m-BMD and DL-BMD in total CT scans (r = 0.961/0.979). The ROC-derived AUC of DL-BMD compared to that of central DXA for the low-BMD and osteoporosis patients was 0.847 and 0.770 respectively. The sensitivity, specificity, and accuracy of DL-BMD compared to central DXA for low BMD were 75.0%, 75.0%, and 75.0%, respectively, and those for osteoporosis were 68.0%, 80.5%, and 77.7%. The AUC of DL-BMD compared to the m-BMD for low BMD and osteoporosis diagnosis were 0.990 and 0.943, respectively. The sensitivity, specificity, and accuracy of DL-BMD compared to m-BMD for low BMD were 95.5%, 93.5%, and 94.6%, and those for osteoporosis were 88.2%, 94.5%, and 92.9%, respectively. DL-BMD exhibited excellent agreement with m-BMD on L1 and L2 vertebrae in the various routine clinical CT scans and had comparable diagnostic performance for detecting the low-BMD and osteoporosis on conventional QCT.

A study on whether deep learning models based on CT images for bone density classification and prediction can be used for opportunistic osteoporosis screening

This study utilized deep learning to classify osteoporosis and predict bone density using opportunistic CT scans and independently tested the models on data from different hospitals and equipment. Results showed high accuracy and strong correlation with QCT results, showing promise for expanding osteoporosis screening and reducing unnecessary radiation and costs.

PURPOSE

To explore the feasibility of using deep learning to establish a model for osteoporosis classification and bone density value prediction based on opportunistic CT scans and to verify its generalization and diagnostic ability using an independent test set.

METHODS

A total of 1219 cases of opportunistic CT scans were included in this study, with QCT results as the reference standard. The training set: test set: independent test set ratio was 703: 176: 340, and the independent test set data of 340 cases were from 3 different hospitals and 4 different CT scanners. The VB-Net structure automatic segmentation model was used to segment the trabecular bone, and DenseNet was used to establish a three-classification model and bone density value prediction regression model. The performance parameters of the models were calculated and evaluated.

RESULTS

The ROC curves showed that the mean AUCs of the three-category classification model for categorizing cases into "normal," "osteopenia," and "osteoporosis" for the training set, test set, and independent test set were 0.999, 0.970, and 0.933, respectively. The F1 score, accuracy, precision, recall, precision, and specificity of the test set were 0.903, 0.909, 0.899, 0.908, and 0.956, respectively, and those of the independent test set were 0.798, 0.815, 0.792, 0.81, and 0.899, respectively. The MAEs of the bone density prediction regression model in the training set, test set, and independent test set were 3.15, 6.303, and 10.257, respectively, and the RMSEs were 4.127, 8.561, and 13.507, respectively. The R-squared values were 0.991, 0.962, and 0.878, respectively. The Pearson correlation coefficients were 0.996, 0.981, and 0.94, respectively, and the p values were all < 0.001. The predicted values and bone density values were highly positively correlated, and there was a significant linear relationship.

CONCLUSION

Using deep learning neural networks to process opportunistic CT scan images of the body can accurately predict bone density values and perform bone density three-classification diagnosis, which can reduce the radiation risk, economic consumption, and time consumption brought by specialized bone density measurement, expand the scope of osteoporosis screening, and have broad application prospects.

Automated vertebral bone mineral density measurement with phantomless internal calibration in chest LDCT scans using deep learning

OBJECTIVE

To develop and evaluate a fully automated method based on deep learning and phantomless internal calibration for bone mineral density (BMD) measurement and opportunistic low BMD (osteopenia and osteoporosis) screening using chest low-dose CT (LDCT) scans.

METHODS

A total of 1175 individuals were enrolled in this study, who underwent both chest LDCT and BMD examinations with quantitative computed tomography (QCT), by two different CT scanners (Siemens and GE). Two convolutional neural network (CNN) models were employed for vertebral body segmentation and labeling, respectively. A histogram technique was applied for vertebral BMD calculation using paraspinal muscle and surrounding fat as references. 195 cases (by Siemens scanner) as fitting cohort were used to build the calibration function. 698 cases as validation cohort I (VCI, by Siemens scanner) and 282 cases as validation cohort II (VCII, by GE scanner) were performed to evaluate the performance of the proposed method, with QCT as the standard for analysis.

RESULTS

The average BMDs from the proposed method were strongly correlated with QCT (in VCI: r = 0.896, in VCII: r = 0.956, p < 0.001). Bland-Altman analysis showed a small mean difference of 1.1 mg/cm3, and large interindividual differences as seen by wide 95% limits of agreement (-29.9 to +32.0 mg/cm3) in VCI. The proposed method measured BMDs were higher than QCT measured BMDs in VCII (mean difference = 15.3 mg/cm3, p < 0.001). Osteoporosis and low BMD were diagnosed by proposed method with AUCs of 0.876 and 0.903 in VCI, 0.731 and 0.794 in VCII, respectively. The AUCs of the proposed method were increased to over 0.920 in both VCI and VCII after adjusting the cut-off.

CONCLUSION

Without manual selection of the region of interest of body tissues, the proposed method based on deep learning and phantomless internal calibration has the potential for preliminary screening of patients with low BMD using chest LDCT scans. However, the agreement between the proposed method and QCT is insufficient to allow them to be used interchangeably in BMD measurement.

ADVANCES IN KNOWLEDGE

This study proposed an automated vertebral BMD measurement method based on deep learning and phantomless internal calibration with paraspinal muscle and fat as reference.

Generalizability of Deep Learning Classification of Spinal Osteoporotic Compression Fractures on Radiographs Using an Adaptation of the Modified-2 Algorithm-Based Qualitative Criteria

RATIONALE AND OBJECTIVES

Spinal osteoporotic compression fractures (OCFs) can be an early biomarker for osteoporosis but are often subtle, incidental, and underreported. To ensure early diagnosis and treatment of osteoporosis, we aimed to build a deep learning vertebral body classifier for OCFs as a critical component of our future automated opportunistic screening tool.

MATERIALS AND METHODS

We retrospectively assembled a local dataset, including 1790 subjects and 15,050 vertebral bodies (thoracic and lumbar). Each vertebral body was annotated using an adaption of the modified-2 algorithm-based qualitative criteria. The Osteoporotic Fractures in Men (MrOS) Study dataset provided thoracic and lumbar spine radiographs of 5994 men from six clinical centers. Using both datasets, five deep learning algorithms were trained to classify each individual vertebral body of the spine radiographs. Classification performance was compared for these models using multiple metrics, including the area under the receiver operating characteristic curve (AUC-ROC), sensitivity, specificity, and positive predictive value (PPV).

RESULTS

Our best model, built with ensemble averaging, achieved an AUC-ROC of 0.948 and 0.936 on the local dataset's test set and the MrOS dataset's test set, respectively. After setting the cutoff threshold to prioritize PPV, this model achieved a sensitivity of 54.5% and 47.8%, a specificity of 99.7% and 99.6%, and a PPV of 89.8% and 94.8%.

CONCLUSION

Our model achieved an AUC-ROC>0.90 on both datasets. This testing shows some generalizability to real-world clinical datasets and a suitable performance for a future opportunistic osteoporosis screening tool.

Artificial Intelligence Applications for Osteoporosis Classification Using Computed Tomography

Osteoporosis, marked by low bone mineral density (BMD) and a high fracture risk, is a major health issue. Recent progress in medical imaging, especially CT scans, offers new ways of diagnosing and assessing osteoporosis. This review examines the use of AI analysis of CT scans to stratify BMD and diagnose osteoporosis. By summarizing the relevant studies, we aimed to assess the effectiveness, constraints, and potential impact of AI-based osteoporosis classification (severity) via CT. A systematic search of electronic databases (PubMed, MEDLINE, Web of Science, ClinicalTrials.gov) was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A total of 39 articles were retrieved from the databases, and the key findings were compiled and summarized, including the regions analyzed, the type of CT imaging, and their efficacy in predicting BMD compared with conventional DXA studies. Important considerations and limitations are also discussed. The overall reported accuracy, sensitivity, and specificity of AI in classifying osteoporosis using CT images ranged from 61.8% to 99.4%, 41.0% to 100.0%, and 31.0% to 100.0% respectively, with areas under the curve (AUCs) ranging from 0.582 to 0.994. While additional research is necessary to validate the clinical efficacy and reproducibility of these AI tools before incorporating them into routine clinical practice, these studies demonstrate the promising potential of using CT to opportunistically predict and classify osteoporosis without the need for DEXA.

Artificial Intelligence and Machine Learning in Lung Cancer Screening

Recent advances in artificial intelligence and machine learning (AI/ML) hold substantial promise to address some of the current challenges in lung cancer screening and improve health equity. This article reviews the status and future directions of AI/ML tools in the lung cancer screening workflow, focusing on determining screening eligibility, radiation dose reduction and image denoising for low-dose chest computed tomography (CT), lung nodule detection, lung nodule classification, and determining optimal screening intervals. AI/ML tools can assess for chronic diseases on CT, which creates opportunities to improve population health through opportunistic screening.

Performance evaluation of a deep learning-based cascaded HRNet model for automatic measurement of X-ray imaging parameters of lumbar sagittal curvature

PURPOSE

To develop a deep learning-based cascaded HRNet model, in order to automatically measure X-ray imaging parameters of lumbar sagittal curvature and to evaluate its prediction performance.

METHODS

A total of 3730 lumbar lateral digital radiography (DR) images were collected from picture archiving and communication system (PACS). Among them, 3150 images were randomly selected as the training dataset and validation dataset, and 580 images as the test dataset. The landmarks of the lumbar curve index (LCI), lumbar lordosis angle (LLA), sacral slope (SS), lumbar lordosis index (LLI), and the posterior edge tangent angle of the vertebral body (PTA) were identified and marked. The measured results of landmarks on the test dataset were compared with the mean values of manual measurement as the reference standard. Percentage of correct key-points (PCK), intra-class correlation coefficient (ICC), Pearson correlation coefficient (r), mean absolute error (MAE), mean square error (MSE), root-mean-square error (RMSE), and Bland-Altman plot were used to evaluate the performance of the cascade HRNet model.

RESULTS

The PCK of the cascaded HRNet model was 97.9-100% in the 3 mm distance threshold. The mean differences between the reference standard and the predicted values for LCI, LLA, SS, LLI, and PTA were 0.43 mm, 0.99°, 1.11°, 0.01 mm, and 0.23°, respectively. There were strong correlation and consistency of the five parameters between the cascaded HRNet model and manual measurements (ICC = 0.989-0.999, R = 0.991-0.999, MAE = 0.63-1.65, MSE = 0.61-4.06, RMSE = 0.78-2.01).

CONCLUSION

The cascaded HRNet model based on deep learning algorithm could accurately identify the sagittal curvature-related landmarks on lateral lumbar DR images and automatically measure the relevant parameters, which is of great significance in clinical application.

Screening for osteoporosis based on IQon spectral CT virtual low monoenergetic images: Comparison with conventional 120 kVp images

Objectives

To explore the differences between low kiloelectron volt (keV) virtual monoenergetic images (VMIs) using IQon spectral CT and conventional CT (120 kVp) in the diagnosis of osteoporosis.

Methods

This retrospective study included 317 patients who underwent IQon spectral CT and dual-energy X-ray absorptiometry (DXA) examination. Commercial deep learning-based software was used for the fully automated extraction of the CT values of the first to fourth lumbar vertebrae (L1-L4) from two different low-keV levels (including 40/70 keV) VMIs and conventional 120 kVp images. The DXA examination results served as the standard of reference (normal [T-score ≥ -1], osteopenia [-2.5 < T-score < -1], and osteoporosis [T-score < -2.5]). Osteoporosis diagnosis models were constructed using machine learning classifiers (logistic regression, support vector machine, random forest, XGBoost, and multilayer perceptron) based on the average CT values of L1-L4. The area under the receiver operating characteristic curve (AUC) and DeLong test were performed to compare differences in the performance of the osteoporosis diagnosis model between virtual low-keV VMIs and standard 120 kVp images.

Results

Random forest-based prediction model obtained good overall performance among all classifiers, and macro/micro average AUC values of 0.820/840, 0.834/853, and 0.831/852 were obtained based on 40/70 keV and 120 kVp images, respectively. The model presented no significant difference between low-keV VMIs and standard 120 kVp images for the diagnosis of osteoporosis (p > 0.05).

Conclusions

The performance of the osteoporosis diagnosis model using IQon spectral CT simulating the low tube voltage scanning condition (less than 120 kVp) was also satisfactory. Bone density screening evaluation can be performed with a combination of low-dose lung scanning CT, greatly reducing the radiation dose without affecting the diagnosis.

Opportunistic AI-enabled automated bone mineral density measurements in lung cancer screening and coronary calcium scoring CT scans are equivalent

Rationale and objectives

We previously reported a novel manual method for measuring bone mineral density (BMD) in coronary artery calcium (CAC) scans and validated our method against Dual X-Ray Absorptiometry (DEXA). Furthermore, we have developed and validated an artificial intelligence (AI) based automated BMD (AutoBMD) measurement as an opportunistic add-on to CAC scans that recently received FDA approval. In this report, we present evidence of equivalency between AutoBMD measurements in cardiac vs lung CT scans.

Materials and methods

AI models were trained using 132 cases with 7649 (3 mm) slices for CAC, and 37 cases with 21918 (0.5 mm) slices for lung scans. To validate AutoBMD against manual measurements, we used 6776 cases of BMD measured manually on CAC scans in the Multi-Ethnic Study of Atherosclerosis (MESA). We then used 165 additional cases from Harbor UCLA Lundquist Institute to compare AutoBMD in patients who underwent both cardiac and lung scans on the same day.

Results

Mean±SD for age was 69 ± 9.4 years with 52.4% male. AutoBMD in lung and cardiac scans, and manual BMD in cardiac scans were 153.7 ± 43.9, 155.1 ± 44.4, and 163.6 ± 45.3 g/cm³, respectively (p = 0.09). Bland-Altman agreement analysis between AutoBMD lung and cardiac scans resulted in 1.37 g/cm³ mean differences. Pearson correlation coefficient between lung and cardiac AutoBMD was R² = 0.95 (p < 0.0001).

Conclusion

Opportunistic BMD measurement using AutoBMD in CAC and lung cancer screening scans is promising and yields similar results. No extra radiation plus the high prevalence of asymptomatic osteoporosis makes AutoBMD an ideal screening tool for osteopenia and osteoporosis in CT scans done for other reasons.

CT-based radiomics can identify physiological modifications of bone structure related to subjects' age and sex

PURPOSE

Radiomics of vertebral bone structure is a promising technique for identification of osteoporosis. We aimed at assessing the accuracy of machine learning in identifying physiological changes related to subjects' sex and age through analysis of radiomics features from CT images of lumbar vertebrae, and define its generalizability across different scanners.

MATERIALS AND METHODS

We annotated spherical volumes-of-interest (VOIs) in the center of the vertebral body for each lumbar vertebra in 233 subjects who had undergone lumbar CT for back pain on 3 different scanners, and we evaluated radiomics features from each VOI. Subjects with history of bone metabolism disorders, cancer, and vertebral fractures were excluded. We performed machine learning classification and regression models to identify subjects' sex and age respectively, and we computed a voting model which combined predictions.

RESULTS

The model was trained on 173 subjects and tested on an internal validation dataset of 60. Radiomics was able to identify subjects' sex within single CT scanner (ROC AUC: up to 0.9714), with lower performance on the combined dataset of the 3 scanners (ROC AUC: 0.5545). Higher consistency among different scanners was found in identification of subjects' age (R2 0.568 on all scanners, MAD 7.232 years), with highest results on a single CT scanner (R2 0.667, MAD 3.296 years).

CONCLUSION

Radiomics features are able to extract biometric data from lumbar trabecular bone, and determine bone modifications related to subjects' sex and age with great accuracy. However, acquisition from different CT scanners reduces the accuracy of the analysis.

New Horizons: Artificial Intelligence Tools for Managing Osteoporosis

Osteoporosis is a disease characterized by low bone mass and microarchitectural deterioration leading to increased bone fragility and fracture risk. Typically, osteoporotic fractures occur at the spine, hip, distal forearm, and proximal humerus, but other skeletal sites may be affected as well. One of the major challenges in the management of osteoporosis lies in the fact that although the operational diagnosis is based on bone mineral density (BMD) as measured by dual x-ray absorptiometry, the majority of fractures occur at nonosteoporotic BMD values. Furthermore, osteoporosis often remains undiagnosed regardless of the low severity of the underlying trauma. Also, there is only weak consensus among the major guidelines worldwide, when to treat, whom to treat, and which drug to use. Against this background, increasing efforts have been undertaken in the past few years by artificial intelligence (AI) developers to support and improve the management of this disease. The performance of many of these newly developed AI algorithms have been shown to be at least comparable to that of physician experts, or even superior. However, even if study results appear promising at a first glance, they should always be interpreted with caution. Use of inadequate reference standards or selection of variables that are of little or no value in clinical practice are limitations not infrequently found. Consequently, there is a clear need for high-quality clinical research in this field of AI. This could, eg, be achieved by establishing an internationally consented "best practice framework" that considers all relevant stakeholders.

Opportunistic screening for low bone density using abdominopelvic computed tomography scans

BACKGROUND

While low bone density is a major burden on US health system, current osteoporosis screening guidelines by the US Preventive Services Task Force are limited to women aged ≥65 and all postmenopausal women with certain risk factors. Even within recommended screening groups, actual screening rates are low (<26%) and vary across socioeconomic groups. The proposed model can opportunistically screen patients using abdominal CT studies for low bone density who may otherwise go undiagnosed.

PURPOSE

To develop an artificial intelligence (AI) model for opportunistic screening of low bone density using both contrast and non-contrast abdominopelvic computed tomography (CT) exams, for the purpose of referral to traditional bone health management, which typically begins with dual energy X-ray absorptiometry (DXA).

METHODS

We collected 6083 contrast-enhanced CT imaging exams paired with DXA exams within ±6 months documented between May 2015 and August 2021 in a single institution with four major healthcare practice regions. Our fusion AI pipeline receives the coronal and axial plane images of a contrast enhanced abdominopelvic CT exam and basic patient demographics (age, gender, body cross section lengths) to predict risk of low bone mass. The models were trained on lumbar spine T-scores from DXA exams and tested on multi-site imaging exams. The model was again tested in a prospective group (N = 344) contrast-enhanced and non-contrast-enhanced studies.

RESULTS

The models were evaluated on the same test set (1208 exams)-(1) Baseline model using demographic factors from electronic medical records (EMR) - 0.7 area under the curve of receiver operator characteristic (AUROC); Imaging based models: (2) axial view - 0.83 AUROC; (3) coronal view- 0.83 AUROC; (4) Fusion model-Imaging + demographic factors - 0.86 AUROC. The prospective test yielded one missed positive DXA case with a hip prosthesis among 23 positive contrast-enhanced CT exams and 0% false positive rate for non-contrast studies. Both positive cases among non-contrast enhanced CT exams were successfully detected. While only about 8% patients from prospective study received a DXA exam within 2 years, about 30% were detected with low bone mass by the fusion model, highlighting the need for opportunistic screening.

CONCLUSIONS

The fusion model, which combines two planes of CT images and EMRs data, outperformed individual models and provided a high, robust diagnostic performance for opportunistic screening of low bone density using contrast and non-contrast CT exams. This model could potentially improve bone health risk assessment with no additional cost. The model's handling of metal implants is an ongoing effort.

Prediction of osteoporosis using radiomics analysis derived from single source dual energy CT

BACKGROUND

With the aging population of society, the incidence rate of osteoporosis is increasing year by year. Early diagnosis of osteoporosis plays a significant role in the progress of disease prevention. As newly developed technology, computed tomography (CT) radiomics could discover radiomic features difficult to recognize visually, providing convenient, comprehensive and accurate osteoporosis diagnosis. This study aimed to develop and validate a clinical-radiomics model based on the monochromatic imaging of single source dual-energy CT for osteoporosis prediction.

METHODS

One hundred sixty-four participants who underwent both single source dual-energy CT and quantitative computed tomography (QCT) lumbar-spine examination were enrolled in a study cohort including training datasets (n = 114 [30 osteoporosis and 84 non-osteoporosis]) and validation datasets (n = 50 [12 osteoporosis and 38 non-osteoporosis]). One hundred seven radiomics features were extracted from 70-keV monochromatic CT images. With QCT as the reference standard, a radiomics signature was built by using least absolute shrinkage and selection operator (LASSO) regression on the basis of reproducible features. A clinical-radiomics model was constructed by incorporating the radiomics signature and a significant clinical predictor (age) using multivariate logistic regression analysis. Model performance was assessed by its calibration, discrimination and clinical usefulness.

RESULTS

The radiomics signature comprised 14 selected features and showed good calibration and discrimination in both training and validation cohorts. The clinical-radiomics model, which incorporated the radiomics signature and a significant clinical predictor (age), also showed good discrimination, with an area under the receiver operating characteristic curve (AUC) of 0.938 (95% confidence interval, 0.903-0.952) in the training cohort and an AUC of 0.988 (95% confidence interval, 0.967-0.998) in the validation cohort, and good calibration. The clinical-radiomics model stratified participants into groups with osteoporosis and non-osteoporosis with an accuracy of 94.0% in the validation cohort. Decision curve analysis (DCA) demonstrated that the radiomics signature and the clinical-radiomics model were clinically useful.

CONCLUSIONS

The clinical-radiomics model incorporating the radiomics signature and a clinical parameter had a good ability to predict osteoporosis based on dual-energy CT monoenergetic imaging.

Interpretable Deep-Learning Approaches for Osteoporosis Risk Screening and Individualized Feature Analysis Using Large Population-Based Data: Model Development and Performance Evaluation

BACKGROUND

Osteoporosis is one of the diseases that requires early screening and detection for its management. Common clinical tools and machine-learning (ML) models for screening osteoporosis have been developed, but they show limitations such as low accuracy. Moreover, these methods are confined to limited risk factors and lack individualized explanation.

OBJECTIVE

The aim of this study was to develop an interpretable deep-learning (DL) model for osteoporosis risk screening with clinical features. Clinical interpretation with individual explanations of feature contributions is provided using an explainable artificial intelligence (XAI) technique.

METHODS

We used two separate data sets: the National Health and Nutrition Examination Survey data sets from the United States (NHANES) and South Korea (KNHANES) with 8274 and 8680 respondents, respectively. The study population was classified according to the T-score of bone mineral density at the femoral neck or total femur. A DL model for osteoporosis diagnosis was trained on the data sets and significant risk factors were investigated with local interpretable model-agnostic explanations (LIME). The performance of the DL model was compared with that of ML models and conventional clinical tools. Additionally, contribution ranking of risk factors and individualized explanation of feature contribution were examined.

RESULTS

Our DL model showed area under the curve (AUC) values of 0.851 (95% CI 0.844-0.858) and 0.922 (95% CI 0.916-0.928) for the femoral neck and total femur bone mineral density, respectively, using the NHANES data set. The corresponding AUC values for the KNHANES data set were 0.827 (95% CI 0.821-0.833) and 0.912 (95% CI 0.898-0.927), respectively. Through the LIME method, significant features were induced, and each feature's integrated contribution and interpretation for individual risk were determined.

CONCLUSIONS

The developed DL model significantly outperforms conventional ML models and clinical tools. Our XAI model produces high-ranked features along with the integrated contributions of each feature, which facilitates the interpretation of individual risk. In summary, our interpretable model for osteoporosis risk screening outperformed state-of-the-art methods.

Medical optimization of osteoporosis for adult spinal deformity surgery: a state-of-the-art evidence-based review of current pharmacotherapy

PURPOSE

Osteoporosis is a common, but challenging phenomenon to overcome in adult spinal deformity (ASD) surgery. Several pharmacological agents are at the surgeon's disposal to optimize the osteoporotic patient prior to undergoing extensive reconstruction. Familiarity with these medications will allow the surgeon to make informed decisions on selecting the most appropriate adjuncts for each individual patient.

METHODS

A comprehensive literature review was conducted in PubMed from September 2021 to April 2022. Studies were selected that contained combinations of various terms including osteoporosis, specific medications, spine surgery, fusion, cage subsidence, screw loosening, pull-out, junctional kyphosis/failure.

RESULTS

Bisphosphonates, denosumab, selective estrogen receptor modulators, teriparatide, abaloparatide and romosozumab are all pharmacological agents currently available for adjunctive use. While these medications have been shown to have beneficial effects on improving bone mineral density in the osteoporotic patient, varying evidence is available on their specific effects in the context of extensive spine surgery. There is still a lack of human studies with use of the newer agents.

CONCLUSION

Bisphosphonates are first-line agents due to their low cost and robust evidence behind their utility. However, in the absence of contraindications, optimizing bone quality with anabolic medications should be strongly considered in preparation for spinal deformity surgeries due to their beneficial and favorable effects on fusion and hardware compared to the anti-resorptive medications.

Deep Learning Classification of Spinal Osteoporotic Compression Fractures on Radiographs using an Adaptation of the Genant Semiquantitative Criteria

RATIONALE AND OBJECTIVES

Osteoporosis affects 9% of individuals over 50 in the United States and 200 million women globally. Spinal osteoporotic compression fractures (OCFs), an osteoporosis biomarker, are often incidental and under-reported. Accurate automated opportunistic OCF screening can increase the diagnosis rate and ensure adequate treatment. We aimed to develop a deep learning classifier for OCFs, a critical component of our future automated opportunistic screening tool.

MATERIALS AND METHODS

The dataset from the Osteoporotic Fractures in Men Study comprised 4461 subjects and 15,524 spine radiographs. This dataset was split by subject: 76.5% training, 8.5% validation, and 15% testing. From the radiographs, 100,409 vertebral bodies were extracted, each assigned one of two labels adapted from the Genant semiquantitative system: moderate to severe fracture vs. normal/trace/mild fracture. GoogLeNet, a deep learning model, was trained to classify the vertebral bodies. The classification threshold on the predicted probability of OCF outputted by GoogLeNet was set to prioritize the positive predictive value (PPV) while balancing it with the sensitivity. Vertebral bodies with the top 0.75% predicted probabilities were classified as moderate to severe fracture.

RESULTS

Our model yielded a sensitivity of 59.8%, a PPV of 91.2%, and an F1 score of 0.72. The areas under the receiver operating characteristic curve (AUC-ROC) and the precision-recall curve were 0.99 and 0.82, respectively.

CONCLUSION

Our model classified vertebral bodies with an AUC-ROC of 0.99, providing a critical component for our future automated opportunistic screening tool. This could lead to earlier detection and treatment of OCFs.

Three-dimensional reconstruction of Kambin's triangle based on automated magnetic resonance image segmentation

The three-dimensional (3D) anatomy of Kambin's triangle is crucial for surgical planning in minimally invasive spine surgery via the transforaminal approach. Few pieces of research have, however, used image segmentation to explore the 3D reconstruction of Kambin's triangle. This study aimed to develop a new method of 3D reconstruction of Kambin's triangle based on automated magnetic resonance image (MRI) segmentation of the lumbar spinal structures. An experienced (>5 years) "ground truth" spinal pain physician meticulously segmented and labeled spinal structures (e.g., bones, dura mater, discs, and nerve roots) on MRI. Subsequently, a 3D U-Net algorithm was developed for automatically segmenting lumbar spinal structures for the 3D reconstruction of Kambin's triangle. The Dice similarity coefficient (DSC), precision, recall, and the area of Kambin's triangle were used to assess anatomical performance. The automatic segmentation of all spinal structures at the L4/L5 levels and L5/S1 levels resulted in good performance: DSC = 0.878/0.883, precision = 0.889/0.890, recall = 0.873/0.882. Furthermore, the area measurements of Kambin's triangle revealed no significant difference between ground truth and automatic segmentation (p = 0.333 at the L4/L5 level, p = 0.302 at the L5/S1 level). The 3D U-Net model used in this study performed well in terms of simultaneous segmentation of multi-class spinal structures (including bones, dura mater, discs, and nerve roots) on MRI, allowing for accurate 3D reconstruction of Kambin's triangle.

Application of Deep Convolutional Neural Networks in the Diagnosis of Osteoporosis

The aim of this study was to assess the possibility of using deep convolutional neural networks (DCNNs) to develop an effective method for diagnosing osteoporosis based on CT images of the spine. The research material included the CT images of L1 spongy tissue belonging to 100 patients (50 healthy and 50 diagnosed with osteoporosis). Six pre-trained DCNN architectures with different topological depths (VGG16, VGG19, MobileNetV2, Xception, ResNet50, and InceptionResNetV2) were used in the study. The best results were obtained for the VGG16 model characterised by the lowest topological depth (ACC = 95%, TPR = 96%, and TNR = 94%). A specific challenge during the study was the relatively small (for deep learning) number of observations (400 images). This problem was solved using DCNN models pre-trained on a large dataset and a data augmentation technique. The obtained results allow us to conclude that the transfer learning technique yields satisfactory results during the construction of deep models for the diagnosis of osteoporosis based on small datasets of CT images of the spine.

The Potential Role of Artificial Intelligence in Lung Cancer Screening Using Low-Dose Computed Tomography

Two large randomized controlled trials of low-dose CT (LDCT)-based lung cancer screening (LCS) in high-risk smoker populations have shown a reduction in the number of lung cancer deaths in the screening group compared to a control group. Even if various countries are currently considering the implementation of LCS programs, recurring doubts and fears persist about the potentially high false positive rates, cost-effectiveness, and the availability of radiologists for scan interpretation. Artificial intelligence (AI) can potentially increase the efficiency of LCS. The objective of this article is to review the performances of AI algorithms developed for different tasks that make up the interpretation of LCS CT scans, and to estimate how these AI algorithms may be used as a second reader. Despite the reduction in lung cancer mortality due to LCS with LDCT, many smokers die of comorbid smoking-related diseases. The identification of CT features associated with these comorbidities could increase the value of screening with minimal impact on LCS programs. Because these smoking-related conditions are not systematically assessed in current LCS programs, AI can identify individuals with evidence of previously undiagnosed cardiovascular disease, emphysema or osteoporosis and offer an opportunity for treatment and prevention.

Osteoporosis diagnosis in knee X-rays by transfer learning based on convolution neural network

Osteoporosis degrades the quality of bones and is the primary cause of fractures in the elderly and women after menopause. The high diagnostic and treatment costs urge the researchers to find a cost-effective diagnostic system to diagnose osteoporosis in the early stages. X-ray imaging is the cheapest and most common imaging technique to detect bone pathologies butmanual interpretation of x-rays for osteoporosis is difficult and extraction of required features and selection of high-performance classifiers is a very challenging task. Deep learning systems have gained the popularity in image analysis field over the last few decades. This paper proposes a convolution neural network (CNN) based approach to detect osteoporosis from x-rays. In our study, we have used the transfer learning of deep learning-based CNNs namely AlexNet, VggNet-16, ResNet, and VggNet -19 to classify the x-ray images of knee joints into normal, osteopenia, and osteoporosis disease groups. The main objectives of the current study are: (i) to present a dataset of 381 knee x-rays medically validated by the T-scores obtained from the Quantitative Ultrasound System, and (ii) to propose a deep learning approach using transfer learning to classify different stages of the disease. The performance of these classifiers is compared and the best accuracy of 91.1% is achieved by pretrained Alexnet architecture on the presented dataset with an error rate of 0.09 and validation loss of 0.54 as compared to the accuracy of 79%, an error rate of 0.21, and validation loss of 0.544 when pretrained network was not used.. The results of the study suggest that a deep learning system with transfer learning can help clinicians to detect osteoporosis in its early stages hence reducing the risk of fractures.

Improved CT-based Osteoporosis Assessment with a Fully Automated Deep Learning Tool

Purpose

To develop, test, and validate a deep learning (DL) tool that improves upon a previous feature-based CT image processing bone mineral density (BMD) algorithm and compare it against the manual reference standard.

Materials and Methods

This single-center, retrospective, Health Insurance Portability and Accountability Act-compliant study included manual L1 trabecular Hounsfield unit measurements from abdominal CT scans in 11 035 patients (mean age, 58 years ± 12 [SD]; 6311 women) as the reference standard. Automated level selection and L1 trabecular region of interest (ROI) placement were then performed in this CT cohort with both a previously validated feature-based image processing tool and a new DL tool. Overall technical success rates and agreement with the manual reference standard were assessed.

Results

The overall success rate of the DL tool in this heterogeneous patient cohort was significantly higher than that of the older image processing BMD algorithm (99.3% vs 89.4%, P < .001). Using this DL tool, the closest median Hounsfield unit values for single-, three-, and seven-slice vertebral ROIs were within 5% of the manual reference standard Hounsfield unit values in 35.1%, 56.9%, and 85.8% of scans; within 10% in 56.6%, 75.6%, and 92.9% of scans; and within 25% in 76.5%, 89.3%, and 97.1% of scans, respectively. Trade-offs in sensitivity and specificity for osteoporosis assessment were observed from the single-slice approach (sensitivity, 39.4%; specificity, 98.3%) to the minimum value of the multislice approach (for seven contiguous slices; sensitivity, 71.3% and specificity, 94.6%).

Conclusion

The new DL BMD tool demonstrated a higher success rate than the older feature-based image processing tool, and its outputs can be targeted for higher specificity or sensitivity for osteoporosis assessment.Keywords: CT, CT-Quantitative, Abdomen/GI, Skeletal-Axial, Spine, Deep Learning, Machine Learning Supplemental material is available for this article. © RSNA, 2022.

The application of artificial intelligence in spine surgery

Due to its obvious advantages in processing big data and image information, the combination of artificial intelligence and medical care may profoundly change medical practice and promote the gradual transition from traditional clinical care to precision medicine mode. In this artical, we reviewed the relevant literatures and found that artificial intelligence was widely used in spine surgery. The application scenarios included etiology, diagnosis, treatment, postoperative prognosis and decision support systems of spinal diseases. The shift to artificial intelligence model in medicine constantly improved the level of doctors' diagnosis and treatment and the development of orthopedics.

Automated Opportunistic Osteoporosis Screening in Routine Computed Tomography of the Spine: Comparison With Dedicated Quantitative CT

Opportunistic osteoporosis screening in nondedicated routine computed tomography (CT) is of increasing importance. The purpose of this study was to compare lumbar volumetric bone mineral density (vBMD) assessed by a convolutional neural network (CNN)-based framework in routine CT to vBMD from dedicated quantitative CT (QCT), and to evaluate the ability of vBMD and surrogate measurements of Hounsfield units (HU) to distinguish between patients with and without osteoporotic vertebral fractures (VFs). A total of 144 patients (median age: 70.7 years, 93 females) with clinical routine CT (eight different CT scanners, 120 kVp or 140 kVp, with and without intravenous contrast medium) and dedicated QCT acquired within ≤30 days were included. Vertebral measurements included (i) vBMD from the CNN-based approach including automated vertebral body labeling, segmentation, and correction of the contrast media phase for routine CT data (vBMD_OPP), (ii) vBMD from dedicated QCT (vBMD_QCT), and (iii) noncalibrated HU from vertebral bodies of routine CT data as previously proposed for immanent opportunistic osteoporosis screening based on CT attenuation. The intraclass correlation coefficient (ICC) for vBMD_QCT versus vBMD_OPP indicated better agreement (ICC = 0.913) than the ICC for vBMD_QCT versus noncalibrated HU (ICC = 0.704). Bland-Altman analysis showed data points from 137 patients (95.1%) within the limits of agreement (LOA) of -23.2 to 25.0 mg/cm³ for vBMD_QCT versus vBMD_OPP. Osteoporosis (vBMD <80 mg/cm³ ) was detected in 89 patients (vBMD_QCT) and 88 patients (vBMD_OPP), whereas no patient crossed the diagnostic thresholds from normal vBMD to osteoporosis or vice versa. In a subcohort of 88 patients (thoracolumbar spine covered by imaging for VF reading), 69 patients showed one or more prevalent VFs, and the performance for discrimination between patients with and without VFs was best for vBMD_OPP (area under the curve [AUC] = 0.862; 95% confidence interval [CI], 0.771-0.953). In conclusion, automated opportunistic osteoporosis screening in routine CT of various scanner setups is feasible and may demonstrate high diagnostic accuracy for prevalent VFs. © 2022 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).

Current development and prospects of deep learning in spine image analysis: a literature review

BACKGROUND AND OBJECTIVE

As the spine is pivotal in the support and protection of human bodies, much attention is given to the understanding of spinal diseases. Quick, accurate, and automatic analysis of a spine image greatly enhances the efficiency with which spine conditions can be diagnosed. Deep learning (DL) is a representative artificial intelligence technology that has made encouraging progress in the last 6 years. However, it is still difficult for clinicians and technicians to fully understand this rapidly evolving field due to the diversity of applications, network structures, and evaluation criteria. This study aimed to provide clinicians and technicians with a comprehensive understanding of the development and prospects of DL spine image analysis by reviewing published literature.

METHODS

A systematic literature search was conducted in the PubMed and Web of Science databases using the keywords "deep learning" and "spine". Date ranges used to conduct the search were from 1 January, 2015 to 20 March, 2021. A total of 79 English articles were reviewed.

KEY CONTENT AND FINDINGS

The DL technology has been applied extensively to the segmentation, detection, diagnosis, and quantitative evaluation of spine images. It uses static or dynamic image information, as well as local or non-local information. The high accuracy of analysis is comparable to that achieved manually by doctors. However, further exploration is needed in terms of data sharing, functional information, and network interpretability.

CONCLUSIONS

The DL technique is a powerful method for spine image analysis. We believe that, with the joint efforts of researchers and clinicians, intelligent, interpretable, and reliable DL spine analysis methods will be widely applied in clinical practice in the future.

Deep learning for screening primary osteopenia and osteoporosis using spine radiographs and patient clinical covariates in a Chinese population

PURPOSE

Many high-risk osteopenia and osteoporosis patients remain undiagnosed. We proposed to construct a convolutional neural network model for screening primary osteopenia and osteoporosis based on the lumbar radiographs, and to compare the diagnostic performance of the CNN model adding the clinical covariates with the image model alone.

METHODS

A total of 6,908 participants were collected for analysis, including postmenopausal women and men aged 50-95 years, who performed conventional lumbar x-ray examinations and dual-energy x-ray absorptiometry (DXA) examinations within 3 months. All participants were divided into a training set, a validation set, test set 1, and test set 2 at a ratio of 8:1:1:1. The bone mineral density (BMD) values derived from DXA were applied as the reference standard. A three-class CNN model was developed to classify the patients into normal BMD, osteopenia, and osteoporosis. Moreover, we developed the models integrating the images with clinical covariates (age, gender, and BMI), and explored whether adding clinical data improves diagnostic performance over the image mode alone. The receiver operating characteristic curve analysis was performed for assessing the model performance.

RESULTS

As for classifying osteoporosis, the model based on the anteroposterior+lateral channel performed best, with the area under the curve (AUC) range from 0.909 to 0.937 in three test cohorts. The models with images alone achieved moderate sensitivity in classifying osteopenia, in which the highest AUC achieved 0.785. The performance of models integrating images with clinical data shows a slight improvement over models with anteroposterior or lateral images input alone for diagnosing osteoporosis, in which the AUC increased about 2%-4%. Regarding categorizing osteopenia and the normal BMD, the proposed models integrating images with clinical data also outperformed the models with images solely.

CONCLUSION

The deep learning-based approach could screen osteoporosis and osteopenia based on lumbar radiographs.

Augmenting Osteoporosis Imaging with Machine Learning

PURPOSE OF REVIEW

In this paper, we discuss how recent advancements in image processing and machine learning (ML) are shaping a new and exciting era for the osteoporosis imaging field. With this paper, we want to give the reader a basic exposure to the ML concepts that are necessary to build effective solutions for image processing and interpretation, while presenting an overview of the state of the art in the application of machine learning techniques for the assessment of bone structure, osteoporosis diagnosis, fracture detection, and risk prediction.

RECENT FINDINGS

ML effort in the osteoporosis imaging field is largely characterized by "low-cost" bone quality estimation and osteoporosis diagnosis, fracture detection, and risk prediction, but also automatized and standardized large-scale data analysis and data-driven imaging biomarker discovery. Our effort is not intended to be a systematic review, but an opportunity to review key studies in the recent osteoporosis imaging research landscape with the ultimate goal of discussing specific design choices, giving the reader pointers to possible solutions of regression, segmentation, and classification tasks as well as discussing common mistakes.

Applications of Machine Learning in Bone and Mineral Research

In this unprecedented era of the overwhelming volume of medical data, machine learning can be a promising tool that may shed light on an individualized approach and a better understanding of the disease in the field of osteoporosis research, similar to that in other research fields. This review aimed to provide an overview of the latest studies using machine learning to address issues, mainly focusing on osteoporosis and fractures. Machine learning models for diagnosing and classifying osteoporosis and detecting fractures from images have shown promising performance. Fracture risk prediction is another promising field of research, and studies are being conducted using various data sources. However, these approaches may be biased due to the nature of the techniques or the quality of the data. Therefore, more studies based on the proposed guidelines are needed to improve the technical feasibility and generalizability of artificial intelligence algorithms.

Automated bone mineral density prediction and fracture risk assessment using plain radiographs via deep learning

Dual-energy X-ray absorptiometry (DXA) is underutilized to measure bone mineral density (BMD) and evaluate fracture risk. We present an automated tool to identify fractures, predict BMD, and evaluate fracture risk using plain radiographs. The tool performance is evaluated on 5164 and 18175 patients with pelvis/lumbar spine radiographs and Hologic DXA. The model is well calibrated with minimal bias in the hip (slope = 0.982, calibration-in-the-large = -0.003) and the lumbar spine BMD (slope = 0.978, calibration-in-the-large = 0.003). The area under the precision-recall curve and accuracy are 0.89 and 91.7% for hip osteoporosis, 0.89 and 86.2% for spine osteoporosis, 0.83 and 95.0% for high 10-year major fracture risk, and 0.96 and 90.0% for high hip fracture risk. The tool classifies 5206 (84.8%) patients with 95% positive or negative predictive value for osteoporosis, compared to 3008 DXA conducted at the same study period. This automated tool may help identify high-risk patients for osteoporosis.

Differential diagnosis of benign and malignant vertebral fracture on CT using deep learning

OBJECTIVES

To evaluate the performance of deep learning using ResNet50 in differentiation of benign and malignant vertebral fracture on CT.

METHODS

A dataset of 433 patients confirmed with 296 malignant and 137 benign fractures was retrospectively selected from our spinal CT image database. A senior radiologist performed visual reading to evaluate six imaging features, and three junior radiologists gave diagnostic prediction. A ROI was placed on the most abnormal vertebrae, and the smallest square bounding box was generated. The input channel into ResNet50 network was 3, including the slice with its two neighboring slices. The diagnostic performance was evaluated using 10-fold cross-validation. After obtaining the malignancy probability from all slices in a patient, the highest probability was assigned to that patient to give the final diagnosis, using the threshold of 0.5.

RESULTS

Visual features such as soft tissue mass and bone destruction were highly suggestive of malignancy; the presence of a transverse fracture line was highly suggestive of a benign fracture. The reading by three radiologists with 5, 3, and 1 year of experience achieved an accuracy of 99%, 95.2%, and 92.8%, respectively. In ResNet50 analysis, the per-slice diagnostic sensitivity, specificity, and accuracy were 0.90, 0.79, and 85%. When the slices were combined to ve per-patient diagnosis, the sensitivity, specificity, and accuracy were 0.95, 0.80, and 88%.

CONCLUSION

Deep learning has become an important tool for the detection of fractures on CT. In this study, ResNet50 achieved good accuracy, which can be further improved with more cases and optimized methods for future clinical implementation.

KEY POINTS

• Deep learning using ResNet50 can yield a high accuracy for differential diagnosis of benign and malignant vertebral fracture on CT. • The per-slice diagnostic sensitivity, specificity, and accuracy were 0.90, 0.79, and 85% in deep learning using ResNet50 analysis. • The slices combined with per-patient diagnostic sensitivity, specificity, and accuracy were 0.95, 0.80, and 88% in deep learning using ResNet50 analysis.

Machine Learning Solutions for Osteoporosis-A Review

Osteoporosis and its clinical consequence, bone fracture, is a multifactorial disease that has been the object of extensive research. Recent advances in machine learning (ML) have enabled the field of artificial intelligence (AI) to make impressive breakthroughs in complex data environments where human capacity to identify high-dimensional relationships is limited. The field of osteoporosis is one such domain, notwithstanding technical and clinical concerns regarding the application of ML methods. This qualitative review is intended to outline some of these concerns and to inform stakeholders interested in applying AI for improved management of osteoporosis. A systemic search in PubMed and Web of Science resulted in 89 studies for inclusion in the review. These covered one or more of four main areas in osteoporosis management: bone properties assessment (n = 13), osteoporosis classification (n = 34), fracture detection (n = 32), and risk prediction (n = 14). Reporting and methodological quality was determined by means of a 12-point checklist. In general, the studies were of moderate quality with a wide range (mode score 6, range 2 to 11). Major limitations were identified in a significant number of studies. Incomplete reporting, especially over model selection, inadequate splitting of data, and the low proportion of studies with external validation were among the most frequent problems. However, the use of images for opportunistic osteoporosis diagnosis or fracture detection emerged as a promising approach and one of the main contributions that ML could bring to the osteoporosis field. Efforts to develop ML-based models for identifying novel fracture risk factors and improving fracture prediction are additional promising lines of research. Some studies also offered insights into the potential for model-based decision-making. Finally, to avoid some of the common pitfalls, the use of standardized checklists in developing and sharing the results of ML models should be encouraged. © 2021 American Society for Bone and Mineral Research (ASBMR).