Radiomics and Machine Learning Can Differentiate Transient Osteoporosis from Avascular Necrosis of the Hip

Computer-aided diagnosis for China-Japan Friendship Hospital classification of necrotic femurs using statistical shape and appearance model based on CT scans

The purpose of this study is to quantify the three-dimensional (3D) structural morphology, bone mineral density (BMD) distribution, and mechanical properties of different China-Japan Friendship Hospital (CJFH) classification types and assist clinicians in classifying necrotic femurs accurately. In this study, 41 cases were classified as types L2 and L3 based on CT images. Then, 3D Statistical Shape and Appearance Models (SSM and SAM) were established, and 80 principal component (PC) modes were extracted from the SSM and SAM as the candidate features. The bone strength of each case was also calculated as the candidate feature using finite element analysis (FEA). Support vector machine (SVM) and Extreme Gradient Boosting (XGBoost) were used to establish 10 machine learning models. Feature selection methods were used to screen the candidate features. The performance of each model was evaluated based on sensitivity, specificity, accuracy, and the area under the receiver operating characteristic (ROC) curve. This resulted in a SVM model for CJFH classification with the performance: accuracy of 87.5%, sensitivity of 85.0%, specificity of 76.0%, and AUC of 94.2%. This study provided effective machine learning models for assisting in diagnosing CJFH types, increasing the objectivity of the diagnosis. They may have great potential for application in clinical assessments of CJFH classification.

Interpretable machine learning and radiomics in hip MRI diagnostics: comparing ONFH and OA predictions to experts

PURPOSE

Distinguishing between Osteonecrosis of the femoral head (ONFH) and Osteoarthritis (OA) can be subjective and vary between users with different backgrounds and expertise. This study aimed to construct and evaluate several Radiomics-based machine learning models using MRI to differentiate between those two disorders and compare their efficacies to those of medical experts.

METHODS

140 MRI scans were retrospectively collected from the electronic medical records. They were split into training and testing sets in a 7:3 ratio. Handcrafted radiomics features were harvested following the careful manual segmentation of the regions of interest (ROI). After thoroughly selecting these features, various machine learning models have been constructed. The evaluation was carried out using receiver operating characteristic (ROC) curves. Then NaiveBayes (NB) was selected to establish our final Radiomics-model as it performed the best. Three users with different expertise and backgrounds diagnosed and labeled the dataset into either OA or ONFH. Their results have been compared to our Radiomics-model.

RESULTS

The amount of handcrafted radiomics features was 1197 before processing; after the final selection, only 12 key features were retained and used. User 1 had an AUC of 0.632 (95% CI 0.4801-0.7843), User 2 recorded an AUC of 0.565 (95% CI 0.4102-0.7196); while User 3 was on top with an AUC of 0.880 (95% CI 0.7753-0.9843). On the other hand, the Radiomics model attained an AUC of 0.971 (95% CI 0.9298-1.0000); showing greater efficacy than all other users. It also demonstrated a sensitivity of 0.937 and a specificity of 0.885. DCA (Decision Curve Analysis displayed that the radiomics-model had a greater clinical benefit in differentiating OA and ONFH.

CONCLUSION

We have successfully constructed and evaluated an interpretable radiomics-based machine learning model that could distinguish between OA and ONFH. This method has the ability to aid both junior and senior medical professionals to precisely diagnose and take prompt treatment measures.

Prediction of femoral head collapse in osteonecrosis using deep learning segmentation and radiomics texture analysis of MRI

BACKGROUND

Femoral head collapse is a critical pathological change and is regarded as turning point in disease progression in osteonecrosis of the femoral head (ONFH). In this study, we aim to build an automatic femoral head collapse prediction pipeline for ONFH based on magnetic resonance imaging (MRI) radiomics.

METHODS

In the segmentation model development dataset, T1-weighted MRI of 222 hips from two hospitals were retrospectively collected and randomly split into training (n = 190) and test (n = 32) sets. In the prognosis prediction model development dataset, 206 hips were also retrospectively collected from two hospitals and divided into training set (n = 155) and external test set (n = 51) according to data source. A deep learning model for automatic lesion segmentation was trained with nnU-Net, from which three-dimensional regions of interest were segmented and a total of 107 radiomics features were extracted. After intra-class correlation coefficients screening, feature correlation coefficient screening and Least Absolute Shrinkage and Selection Operator regression feature selection, a machine learning model for ONFH prognosis prediction was trained with Logistic Regression (LR) and Light Gradient Boosting Machine (LightGBM) algorithm.

RESULTS

The segmentation model achieved an average dice similarity coefficient of 0.848 and an average 95% Hausdorff distance of 3.794 in the test set, compared to the manual segmentation results. After feature selection, nine radiomics features were included in the prognosis prediction model. External test showed that the LightGBM model exhibited acceptable predictive performance. The area under the curve (AUC) of the prediction model was 0.851 (95% CI: 0.7268-0.9752), with an accuracy of 0.765, sensitivity of 0.833, and specificity of 0.727. Decision curve analysis showed that the LightGBM model exhibited favorable clinical utility.

CONCLUSION

This study presents an automated pipeline for predicting femoral head collapse in ONFH with acceptable performance. Further research is necessary to determine the clinical applicability of this radiomics-based approach and to assess its potential to assist in treatment decision-making for ONFH.

Multimodal radiomics and deep learning models for predicting early femoral head deformity in LCPD

PURPOSE

To develop a predictive model combining clinical, radiomic, and deep learning features based on X-ray and MRI to identify risk factors for early femoral head deformity in Legg-Calvé-Perthes disease (LCPD).

METHODS

This study involved 152 patients diagnosed with early unilateral LCPD across two centers between January 2013 and December 2023, and included an independent external validation set to assess generalizability. Four machine learning methods, including logistic regression (LR), random forest (RF), support vector machine (SVM), and extreme gradient boosting (XGBoost), were employed to develop radiomics deep learning signatures. The clinical-radiomics model (Clinic + Rad), clinical-deep learning model (Clinic + DL), and clinical-radiomics-deep learning model (Clinic + Rad + DL) were developed by integrating radiomics deep learning signatures with clinical variables. The best model, integrated into a nomogram for clinical application, was evaluated using the area under the receiver operating characteristic curve (AUC).

RESULTS

Among the four machine learning methods, XGBoost demonstrated superior performance in our patient dataset: radiomic (Rad) model (AUC, 0.786) and deep learning (DL) model (AUC, 0.803). Clinical variables such as age at onset and JIC classification were associated with early femoral head deformity (p < 0.05). The combined model incorporating clinical, radiomic, and deep learning signatures demonstrated better predictive ability (AUC, 0.853). The nomogram can assist clinicians in effectively assessing the risk of early femoral head deformity.

CONCLUSION

The Clinic + Rad + DL integrated model may be beneficial for prognostic assessment of early LCPD femoral head deformity, which is crucial for tailoring personalized treatment strategies for individual patients.

Prediction of intraoperative press-fit stability of the acetabular cup in total hip arthroplasty using radiomics-based machine learning models

BACKGROUND

Preoperative prediction of the acetabular cup press-fit stability in total hip arthroplasty is necessary for clinical decision-making. This study aims to establish and validate machine learning models to investigate the feasibility of predicting the intraoperative press-fit stability of the acetabular cup in total hip arthroplasty (THA).

METHODS

226 patients who underwent primary THA from 2018 to 2022 in our hospital were retrospectively enrolled. Patients were divided into press-fit stable or unstable groups according to the intraoperative pull-out test of the implanted cup. Then, they were randomly assigned to the training or test cohort in an 8:2 ratio. We used 3Dslicer software to segment the region of interest (ROI) of the patient's bilateral hip X-ray to extract radiomics features. The least absolute shrinkage and selection operator (LASSO) regression was used in our feature selection. Finally, four machine learning models were employed in this study, including support vector machine (SVM), random forest (RF), logistic regression (LR), and XGBoost (XGB). Decision curve analysis (DCA), and receiver operating characteristic (ROC) curves of the models were plotted. The area under the curve (AUC), diagnostic accuracy, sensitivity, and specificity were calculated as well. The AUCs of the four models were compared using the DeLong test.

RESULTS

Twenty-seven valuable radiomics features were determined by dimensionality reduction and selection. Regarding to the DeLong test, the AUC of the XGB model was significantly different from those of the other three models. (p < 0.05). Among all models, the XGB model exhibited the best performance with an AUC of 0.823 (95 % CI: 0.711-0.919) in the test cohort and showed optimal clinical efficacy according to the DCA.

CONCLUSION

Machine learning models based on X-ray radiomics can accurately predict the intraoperative press-fit stability of implanted cups preoperatively, providing surgeons with valuable information to lower the complication risk in THA.

A single sequence MRI-based deep learning radiomics model in the diagnosis of early osteonecrosis of femoral head

Purpose

The objective of this study was to create and assess a Deep Learning-Based Radiomics model using a single sequence MRI that could accurately predict early Femoral Head Osteonecrosis (ONFH). This is the first time such a model was used for the diagnosis of early ONFH. Its simpler than the previously published multi-sequence MRI radiomics based method, and it implements Deep learning to improve on radiomics. It has the potential to be highly beneficial in the early stages of diagnosis and treatment planning.

Methods

MRI scans from 150 patients in total (80 healthy, 70 necrotic) were used, and split into training and testing sets in a 7:3 ratio. Handcrafted as well as deep learning features were retrieved from Tesla 2 weighted (T2W1) MRI slices. After a rigorous selection process, these features were used to construct three models: a Radiomics-based (Rad-model), a Deep Learning-based (DL-model), and a Deep Learning-based Radiomics (DLR-model). The performance of these models in predicting early ONFH was evaluated by comparing them using the receiver operating characteristic (ROC) and decision curve analysis (DCA).

Results

1,197 handcrafted radiomics and 512 DL features were extracted then processed; after the final selection: 15 features were used for the Rad-model, 12 features for the DL-model, and only 9 features were selected for the DLR-model. The most effective algorithm that was used in all of the models was Logistic regression (LR). The Rad-model depicted good results outperforming the DL-model; AUC = 0.944 (95%CI, 0.862-1.000) and AUC = 0.930 (95%CI, 0.838-1.000) respectively. The DLR-model showed superior results to both Rad-model and the DL-model; AUC = 0.968 (95%CI, 0.909-1.000); and a sensitivity of 0.95 and specificity of 0.920. The DCA showed that DLR had a greater net clinical benefit in detecting early ONFH.

Conclusion

Using a single sequence MRI scan, our work constructed and verified a Deep Learning-Based Radiomics Model for early ONFH diagnosis. This strategy outperformed a Deep learning technique based on Resnet18 and a model based on Radiomics. This straightforward method can offer essential diagnostic data promptly and enhance early therapy strategizing for individuals with ONFH, all while utilizing just one MRI sequence and a more standardized and objective interpretation of MRI images.

Multi-sequence MRI-based radiomics: An objective method to diagnose early-stage osteonecrosis of the femoral head

OBJECTIVES

This study investigated the use of radiomics for diagnosing early-stage osteonecrosis of the femoral head (ONFH) by extracting features from multiple MRI sequences and constructing predictive models.

MATERIALS AND METHODS

We conducted a retrospective review, collected MR images of early-stage ONFH (102 from institution A and 20 from institution B) and healthy femoral heads (102 from institution A and 20 from institution B) from two institutions. We extracted radiomics features, handled batch effects using Combat, and normalized features using z-score. We employed the Least absolute shrinkage and selection operator (LASSO) algorithm, along with Max-Relevance and Min-Redundancy (mRMR), to select optimal features for constructing radiomics models based on single, double, and multi-sequence MRI data. We evaluated performance using receiver operating characteristic (ROC) and precision-recall (PR) curves, and compared area under curve of ROC (AUC-ROC) values with the DeLong test. Additionally, we studied the diagnostic performance of the multi-sequence radiomics model and radiologists, compared the diagnostic outcomes of the model and radiologists using the Fisher exact test.

RESULTS

We studied 122 early-stage ONFH and 122 normal femoral heads. The multi-sequence model exhibited the best diagnostic performance among all models (AUC-ROC, PR-AUC for training set: 0.96, 0.961; validation set: 0.96, 0.97; test set: 0.94, 0.94), and it outperformed three resident radiologists on the external testing group with an accuracy of 87.5 %, sensitivity of 85.00 %, and specificity of 90.00 % (p < 0.01), highlighting the robustness of our findings.

CONCLUSIONS

Our study underscored the novelty of the multi-sequence radiomics model in diagnosing early-stage ONFH. By leveraging features extracted from multiple imaging sequences, this approach demonstrated high efficacy, indicating its potential to advance early diagnosis for ONFH. These findings provided important guidance for enhancing early diagnosis of ONFH through radiomics methods, offering new avenues and possibilities for clinical practice and patient care.

Deep learning approach to femoral AVN detection in digital radiography: differentiating patients and pre-collapse stages

OBJECTIVE

This study aimed to evaluate a new deep-learning model for diagnosing avascular necrosis of the femoral head (AVNFH) by analyzing pelvic anteroposterior digital radiography.

METHODS

The study sample included 1167 hips. The radiographs were independently classified into 6 stages by a radiologist using their simultaneous MRIs. After that, the radiographs were given to train and test the deep learning models of the project including SVM and ANFIS layer using the Python programming language and TensorFlow library. In the last step, the test set of hip radiographs was provided to two independent radiologists with different work experiences to compare their diagnosis performance to the deep learning models' performance using the F1 score and Mcnemar test analysis.

RESULTS

The performance of SVM for AVNFH detection (AUC = 82.88%) was slightly higher than less experienced radiologists (79.68%) and slightly lower than experienced radiologists (88.4%) without reaching significance (p-value > 0.05). Evaluation of the performance of SVM for pre-collapse AVNFH detection with an AUC of 73.58% showed significantly higher performance than less experienced radiologists (AUC = 60.70%, p-value < 0.001). On the other hand, no significant difference is noted between experienced radiologists and SVM for pre-collapse detection. ANFIS algorithm for AVNFH detection with an AUC of 86.60% showed significantly higher performance than less experienced radiologists (AUC = 79.68%, p-value = 0.04). Although reaching less performance compared to experienced radiologists statistically not significant (AUC = 88.40%, p-value = 0.20).

CONCLUSIONS

Our study has shed light on the remarkable capabilities of SVM and ANFIS as diagnostic tools for AVNFH detection in radiography. Their ability to achieve high accuracy with remarkable efficiency makes them promising candidates for early detection and intervention, ultimately contributing to improved patient outcomes.

A Case of Bilateral Transient Pregnancy-Related Osteoporosis of the Hip Diagnosed Late During the Lactation Period: A Rare Clinical Presentation and a Mini Review of the Literature

Transient pregnancy-related osteoporosis of the hip is a rare, idiopathic, benign, and usually self-limiting condition caused by edema of the bone marrow, which can be visualized on magnetic resonance imaging. Bilateral localization of the disease is even less common. Our case concerns a 31-year-old primigravida who, during the 35th week of pregnancy, was hospitalized at the Obstetrics and Gynecology Clinic of the General Hospital of Trikala with lumbar and hip pain. The pain gradually increased in intensity and was accompanied by severe movement limitation. No history of falls or injury was reported. Her personal history was unremarkable, and the course of the pregnancy was uneventful. A clinical examination by a team of orthopedic surgeons established a diagnosis of acute hip and back pain. Rest and administration of paracetamol did not improve her clinical condition. During the postpartum and lactation period, the lack of symptom relief led to the decision to further evaluate the patient. The diagnosis of pregnancy-related transient osteoporosis of both hips was established by magnetic resonance imaging. Immediate treatment with bisphosphonate medication after the discontinuation of breastfeeding led to a definitive remission of the symptoms three months later. In this study, after the case description, a brief literature review of this rare clinical entity is presented. Proper knowledge of this condition helps to provide the best possible short- and long-term prognostic outcomes for the mother, fetus, and newborn.

Application of radiomics model based on lumbar computed tomography in diagnosis of elderly osteoporosis

A metabolic bone disease characterized by decreased bone formation and increased bone resorption is osteoporosis. It can cause pain and fracture of patients. The elderly are prone to osteoporosis and are more vulnerable to osteoporosis. In this study, radiomics are extracted from computed tomography (CT) images to screen osteoporosis in the elderly. Collect the plain scan CT images of lumbar spine, cut the region of interest of the image and extract radiomics features, use Lasso regression to screen variables and adjust complexity, use python language to model random forests, support vector machines, K nearest neighbor, and finally use receiver operating characteristic curve to evaluate the performance of the model, including precision, recall, accuracy and area under the curve (AUC). For the model, 14 radiolomics features were selected. The diagnosis performance of random forest model and support vector machine is good, all around 0.9. The AUC of K nearest neighbor model in training set and test set is 0.828 and 0.796, respectively. We selected the plain scan CT images of the elderly lumbar spine to build radiomics features model, which has good diagnostic performance and can be used as a tool to assist the diagnosis of osteoporosis in the elderly.

Deep learning enables the differentiation between early and late stages of hip avascular necrosis

OBJECTIVES

To develop a deep learning methodology that distinguishes early from late stages of avascular necrosis of the hip (AVN) to determine treatment decisions.

METHODS

Three convolutional neural networks (CNNs) VGG-16, Inception ResnetV2, InceptionV3 were trained with transfer learning (ImageNet) and finetuned with a retrospectively collected cohort of (n = 104) MRI examinations of AVN patients, to differentiate between early (ARCO 1-2) and late (ARCO 3-4) stages. A consensus CNN ensemble decision was recorded as the agreement of at least two CNNs. CNN and ensemble performance was benchmarked on an independent cohort of 49 patients from another country and was compared to the performance of two MSK radiologists. CNN performance was expressed with areas under the curve (AUC), the respective 95% confidence intervals (CIs) and precision, and recall and f1-scores. AUCs were compared with DeLong's test.

RESULTS

On internal testing, Inception-ResnetV2 achieved the highest individual performance with an AUC of 99.7% (95%CI 99-100%), followed by InceptionV3 and VGG-16 with AUCs of 99.3% (95%CI 98.4-100%) and 97.3% (95%CI 95.5-99.2%) respectively. The CNN ensemble the same AUCs Inception ResnetV2. On external validation, model performance dropped with VGG-16 achieving the highest individual AUC of 78.9% (95%CI 51.6-79.6%) The best external performance was achieved by the model ensemble with an AUC of 85.5% (95%CI 72.2-93.9%). No significant difference was found between the CNN ensemble and expert MSK radiologists (p = 0.22 and 0.092 respectively).

CONCLUSION

An externally validated CNN ensemble accurately distinguishes between the early and late stages of AVN and has comparable performance to expert MSK radiologists.

CLINICAL RELEVANCE STATEMENT

This paper introduces the use of deep learning for the differentiation between early and late avascular necrosis of the hip, assisting in a complex clinical decision that can determine the choice between conservative and surgical treatment.

KEY POINTS

• A convolutional neural network ensemble achieved excellent performance in distinguishing between early and late avascular necrosis. • The performance of the deep learning method was similar to the performance of expert readers.

The correlation between transient osteoporosis of the hip and pregnancy: A review

Transient osteoporosis of the hip is indubitably a comparatively infrequent entity affecting both men and women worldwide. Its occurrence in the course of pregnancy, specifically in the third trimester, and lactation are of paramount concernment. The exact association between transient hip osteoporosis and pregnancy is precarious. Etiology and potential pathophysiological mechanisms behind this correlation are still to be utterly defined. Magnetic resonance imaging is highly regarded as the gold standard imaging method for assiduous assessment of this disorder. Physicians of copious medical specialties should practice scrupulous techniques for early and pertinent diagnosis when pregnant women are presented with persistent hip pain, as differential diagnosis with femoral head avascular necrosis can be exceedingly arduous. Treatment is predominantly conservative with protected weight-bearing and analgesic medication in the first line of management. In terms of prognosis, the disease ordinarily resolves spontaneously after a few months. Further research is required in order to elucidate the ambiguity surrounding the establishment of globally approved diagnosis and treatment guidelines for pregnancy-associated transient hip osteoporosis. This paper aims to accentuate the significance of this particular disorder by providing a succinct review of the existing literature, augmenting clinicians' knowledge about the features of pregnancy-related transient proximal femur osteoporosis.

Identification of radiomic biomarkers in a set of four skeletal muscle groups on Dixon MRI of the NAKO MR study

In this work, we propose a processing pipeline for the extraction and identification of meaningful radiomics biomarkers in skeletal muscle tissue as displayed using Dixon-weighted MRI. Diverse and robust radiomics features can be identified that may be of aid in the accurate quantification e.g. varying degrees of sarcopenia in respective muscles of large cohorts. As such, the approach comprises the texture feature extraction from raw data based on well established approaches, such as a nnU-Net neural network and the Pyradiomics toolbox, a subsequent selection according to adequate conditions for the muscle tissue of the general population, and an importance-based ranking to further narrow the amount of meaningful features with respect to auxiliary targets. The performance was investigated with respect to the included auxiliary targets, namely age, body mass index (BMI), and fat fraction (FF). Four skeletal muscles with different fiber architecture were included: the mm. glutaei, m. psoas, as well as the extensors and adductors of the thigh. The selection allowed for a reduction from 1015 available texture features to 65 for age, 53 for BMI, and 36 for FF from the available fat/water contrast images considering all muscles jointly. Further, the dependence of the importance rankings calculated for the auxiliary targets on validation sets (in a cross-validation scheme) was investigated by boxplots. In addition, significant differences between subgroups of respective auxiliary targets as well as between both sexes were shown to be present within the ten lowest ranked features by means of Kruskal-Wallis H-tests and Mann-Whitney U-tests. The prediction performance for the selected features and the ranking scheme were verified on validation sets by a random forest based multi-class classification, with strong area under the curve (AUC) values of the receiver operator characteristic (ROC) of 73.03 ± 0.70 % and 73.63 ± 0.70 % for the water and fat images in age, 80.68 ± 0.30 % and 88.03 ± 0.89 % in BMI, as well as 98.36 ± 0.03 % and 98.52 ± 0.09 % in FF.

Radiomics for the Detection of Active Sacroiliitis Using MR Imaging

Detecting active inflammatory sacroiliitis at an early stage is vital for prescribing medications that can modulate disease progression and significantly delay or prevent debilitating forms of axial spondyloarthropathy. Conventional radiography and computed tomography offer limited sensitivity in detecting acute inflammatory findings as these methods primarily identify chronic structural lesions. Conversely, Magnetic Resonance Imaging (MRI) is the preferred technique for detecting bone marrow edema, although it is a complex process requiring extensive expertise. Additionally, ascertaining the origin of lesions can be challenging, even for experienced medical professionals. Machine learning (ML) has showcased its proficiency in various fields by uncovering patterns that are not easily perceived from multi-dimensional datasets derived from medical imaging. The aim of this study is to develop a radiomic signature to aid clinicians in diagnosing active sacroiliitis. A total of 354 sacroiliac joints were segmented from axial fluid-sensitive MRI images, and their radiomic features were extracted. After selecting the most informative features, a number of ML algorithms were utilized to identify the optimal method for detecting active sacroiliitis, leading to the selection of an Extreme Gradient Boosting (XGBoost) model that accomplished an Area Under the Receiver-Operating Characteristic curve (AUC-ROC) of 0.71, thus further showcasing the potential of radiomics in the field.

A pilot radiometabolomics integration study for the characterization of renal oncocytic neoplasia

Differentiating benign renal oncocytic tumors and malignant renal cell carcinoma (RCC) on imaging and histopathology is a critical problem that presents an everyday clinical challenge. This manuscript aims to demonstrate a novel methodology integrating metabolomics with radiomics features (RF) to differentiate between benign oncocytic neoplasia and malignant renal tumors. For this purpose, thirty-three renal tumors (14 renal oncocytic tumors and 19 RCC) were prospectively collected and histopathologically characterised. Matrix-assisted laser desorption/ionisation mass spectrometry imaging (MALDI-MSI) was used to extract metabolomics data, while RF were extracted from CT scans of the same tumors. Statistical integration was used to generate multilevel network communities of -omics features. Metabolites and RF critical for the differentiation between the two groups (delta centrality > 0.1) were used for pathway enrichment analysis and machine learning classifier (XGboost) development. Receiver operating characteristics (ROC) curves and areas under the curve (AUC) were used to assess classifier performance. Radiometabolomics analysis demonstrated differential network node configuration between benign and malignant renal tumors. Fourteen nodes (6 RF and 8 metabolites) were crucial in distinguishing between the two groups. The combined radiometabolomics model achieved an AUC of 86.4%, whereas metabolomics-only and radiomics-only classifiers achieved AUC of 72.7% and 68.2%, respectively. Analysis of significant metabolite nodes identified three distinct tumour clusters (malignant, benign, and mixed) and differentially enriched metabolic pathways. In conclusion, radiometabolomics integration has been presented as an approach to evaluate disease entities. In our case study, the method identified RF and metabolites important in differentiating between benign oncocytic neoplasia and malignant renal tumors, highlighting pathways differentially expressed between the two groups. Key metabolites and RF identified by radiometabolomics can be used to improve the identification and differentiation between renal neoplasms.

Radiomics method in the differential diagnosis of diabetic foot osteomyelitis and charcot neuroarthropathy

OBJECTIVES

Our study used a radiomics method to differentiate bone marrow signal abnormality (BMSA) between Charcot neuroarthropathy (CN) and osteomyelitis (OM).

METHODS AND MATERIALS

The records of 166 patients with diabetic foot suspected CN or OM between January 2020 and March 2022 were retrospectively examined. A total of 41 patients with BMSA on MRI were included in this study. The diagnosis of OM was confirmed histologically in 24 of 41 patients. We clinically followed 17 patients as CN with laboratory tests. We also included 29 nondiabetic patients with traumatic (TR) BMSA on MRI as the third group. Contours of all BMSA on T 2 - and T1 -weighted images in three patient groups were segmented semi-automatically on ManSeg (v.2.7d). The T1 and T2 features of three groups in radiomics were statistically evaluated. We applied the multi-class classification (MCC) and binary-class classification (BCC) methodologies to compare results.

RESULTS

For MCC, the accuracy of Multi-Layer Perceptron (MLP) was 76.92% and 84.38% for T1 and T2, respectively. According to BCC, for CN, OM, and TR BMSA, the sensitivity of MLP is 74%, 89.23%, and 76.19% for T1, and 90.57%, 85.92%, 86.81% for T2, respectively. For CN, OM, and TR BMSA, the specificity of MLP is 89.16%, 87.57%, and 90.72% for T1 and 93.55%, 89.94%, and 90.48% for T2 images, respectively.

CONCLUSION

In diabetic foot, the radiomics method can differentiate the BMSA of CN and OM with high accuracy.

ADVANCES IN KNOWLEDGE

The radiomics method can differentiate the BMSA of CN and OM with high accuracy.

Machine learning algorithms for diagnosis of hip bone osteoporosis: a systematic review and meta-analysis study

BACKGROUND

Osteoporosis is a significant health problem in the skeletal system, associated with bone tissue changes and its strength. Machine Learning (ML), on the other hand, has been accompanied by improvements in recent years and has been in the spotlight. This study is designed to investigate the Diagnostic Test Accuracy (DTA) of ML to detect osteoporosis through the hip dual-energy X-ray absorptiometry (DXA) images.

METHODS

The ISI Web of Science, PubMed, Scopus, Cochrane Library, IEEE Xplore Digital Library, CINAHL, Science Direct, PROSPERO, and EMBASE were systematically searched until June 2023 for studies that tested the diagnostic precision of ML model-assisted for predicting an osteoporosis diagnosis.

RESULTS

The pooled sensitivity of univariate analysis of seven studies was 0.844 (95% CI 0.791 to 0.885, I2 = 94% for 7 studies). The pooled specificity of univariate analysis was 0.781 (95% CI 0.732 to 0.824, I2 = 98% for 7 studies). The pooled diagnostic odds ratio (DOR) was 18.91 (95% CI 14.22 to 25.14, I2 = 93% for 7 studies). The pooled mean positive likelihood ratio (LR+) and the negative likelihood ratio (LR-) were 3.7 and 0.22, respectively. Also, the summary receiver operating characteristics (sROC) of the bivariate model has an AUC of 0.878.

CONCLUSION

Osteoporosis can be diagnosed by ML with acceptable accuracy, and hip fracture prediction was improved via training in an Architecture Learning Network (ALN).

Machine Learning Integrating 99mTc Sestamibi SPECT/CT and Radiomics Data Achieves Optimal Characterization of Renal Oncocytic Tumors

The increasing evidence of oncocytic renal tumors positive in ^99mTc Sestamibi Single Photon Emission Tomography/Computed Tomography (SPECT/CT) examination calls for the development of diagnostic tools to differentiate these tumors from more aggressive forms. This study combined radiomics analysis with the uptake of ^99mTc Sestamibi on SPECT/CT to differentiate benign renal oncocytic neoplasms from renal cell carcinoma. A total of 57 renal tumors were prospectively collected. Histopathological analysis and radiomics data extraction were performed. XGBoost classifiers were trained using the radiomics features alone and combined with the results from the visual evaluation of ^99mTc Sestamibi SPECT/CT examination. The combined SPECT/radiomics model achieved higher accuracy (95%) with an area under the curve (AUC) of 98.3% (95% CI 93.7-100%) than the radiomics-only model (71.67%) with an AUC of 75% (95% CI 49.7-100%) and visual evaluation of ^99mTc Sestamibi SPECT/CT alone (90.8%) with an AUC of 90.8% (95%CI 82.5-99.1%). The positive predictive values of SPECT/radiomics, radiomics-only, and ^99mTc Sestamibi SPECT/CT-only models were 100%, 85.71%, and 85%, respectively, whereas the negative predictive values were 85.71%, 55.56%, and 94.6%, respectively. Feature importance analysis revealed that ^99mTc Sestamibi uptake was the most influential attribute in the combined model. This study highlights the potential of combining radiomics analysis with ^99mTc Sestamibi SPECT/CT to improve the preoperative characterization of benign renal oncocytic neoplasms. The proposed SPECT/radiomics classifier outperformed the visual evaluation of ^99mTc Sestamibii SPECT/CT and the radiomics-only model, demonstrating that the integration of ^99mTc Sestamibi SPECT/CT and radiomics data provides improved diagnostic performance, with minimal false positive and false negative results.

XGBoost-based multiparameters from dual-energy computed tomography for the differentiation of multiple myeloma of the spine from vertebral osteolytic metastases

OBJECTIVES

To evaluate the performance of extreme gradient boosting (XGBoost) combined with multiparameters from dual-energy computed tomography (mpDECT) to differentiate between multiple myeloma (MM) of the spine and vertebral osteolytic metastases (VOM).

METHODS

For this retrospective study, 28 patients (83 lesions) with MM of the spine and 23 patients (54 lesions) with VOM who underwent DECT were included. The mpDECT for each lesion, including normalized effective atomic number, slope of the spectral Hounsfield unit curve, CT attenuation, and virtual noncalcium (VNCa), was obtained. Boruta was used to select the key parameters, and then subsequently merged with XGBoost to yield a prediction model. The lesions were divided into the training and testing group in a 3:1 ratio. The highest performance of the univariate analysis was compared with XGBoost using the Delong test.

RESULTS

The mpDECT of MM was significantly lower than that of VOM (all p < 0.05). In univariate analysis, VNCa had the highest area under the receiver operating characteristic curve (AUC) in the training group (0.81) and testing group (0.87). Based on Boruta, 6 parameters of DECT were selected for XGBoost model construction. The XGBoost model achieved an excellent and stable diagnostic performance, as shown in the training group (AUC of 1.0) and testing group (AUC of 0.97), with a sensitivity of 80%, a specificity of 95%, and an accuracy of 88%, which was superior to VNCa (p < 0.05).

CONCLUSIONS

XGBoost combined with mpDECT yielded promising performance in differentiating between MM of the spine and VOM.

KEY POINTS

• The multiparameters obtained from dual-energy CT of multiple myeloma differed significantly from those of vertebral osteolytic metastases. • The virtual noncalcium offered the highest AUC in the univariate analysis to distinguish multiple myeloma from vertebral osteolytic metastases. • Extreme gradient boosting combined with multiparameters from dual-energy CT had a promising performance to distinguish multiple myeloma from vertebral osteolytic metastases.

Bone marrow lesions of the femoral head: can radiomics distinguish whether it is reversible?

Purpose

Contrary to the self-limiting nature of reversible bone marrow lesions, irreversible bone marrow lesions require early surgical intervention to prevent further morbidity. Thus, early discrimination of irreversible pathology is necessitated. The purpose of this study is to evaluate the efficacy of radiomics and machine learning regarding this topic.

Material and methods

A database was scanned for patients who had undergone MRI of the hip for differential diagnosis of bone marrow lesions and had had follow-up images acquired within 8 weeks after the first imaging. Images that showed resolution of oedema were included in the reversible group. The remainders that showed progression into characteristic signs of osteonecrosis were included in the irreversible group. Radiomics was performed on the first MR images, calculating first- and second-order parameters. Support vector machine and random forest classifiers were performed using these parameters.

Results

Thirty-seven patients (seventeen osteonecrosis) were included. A total of 185 ROIs were segmented. Fortyseven parameters were accepted as classifiers with an area under the curve value ranging from 0.586 to 0.718. Support vector machine yielded a sensitivity of 91.3% and a specificity of 85.1%. Random forest classifier yielded a sensitivity of 84.8% and a specificity of 76.7%. Area under curves were 0.921 for support vector machine and 0.892 for random forest classifier.

Conclusions

Radiomics analysis could prove useful for discrimination of reversible and irreversible bone marrow lesions before the irreversible changes occur, which could prevent morbidities of osteonecrosis by guiding the decisionmaking process for management.

Radiomics and Deep Learning for Disease Detection in Musculoskeletal Radiology: An Overview of Novel MRI- and CT-Based Approaches

ABSTRACT

Radiomics and machine learning-based methods offer exciting opportunities for improving diagnostic performance and efficiency in musculoskeletal radiology for various tasks, including acute injuries, chronic conditions, spinal abnormalities, and neoplasms. While early radiomics-based methods were often limited to a smaller number of higher-order image feature extractions, applying machine learning-based analytic models, multifactorial correlations, and classifiers now permits big data processing and testing thousands of features to identify relevant markers. A growing number of novel deep learning-based methods describe magnetic resonance imaging- and computed tomography-based algorithms for diagnosing anterior cruciate ligament tears, meniscus tears, articular cartilage defects, rotator cuff tears, fractures, metastatic skeletal disease, and soft tissue tumors. Initial radiomics and deep learning techniques have focused on binary detection tasks, such as determining the presence or absence of a single abnormality and differentiation of benign versus malignant. Newer-generation algorithms aim to include practically relevant multiclass characterization of detected abnormalities, such as typing and malignancy grading of neoplasms. So-called delta-radiomics assess tumor features before and after treatment, with temporal changes of radiomics features serving as surrogate markers for tumor responses to treatment. New approaches also predict treatment success rates, surgical resection completeness, and recurrence risk. Practice-relevant goals for the next generation of algorithms include diagnostic whole-organ and advanced classification capabilities. Important research objectives to fill current knowledge gaps include well-designed research studies to understand how diagnostic performances and suggested efficiency gains of isolated research settings translate into routine daily clinical practice. This article summarizes current radiomics- and machine learning-based magnetic resonance imaging and computed tomography approaches for musculoskeletal disease detection and offers a perspective on future goals and objectives.

Deep Learning for the Differential Diagnosis between Transient Osteoporosis and Avascular Necrosis of the Hip

Differential diagnosis between avascular necrosis (AVN) and transient osteoporosis of the hip (TOH) can be complicated even for experienced MSK radiologists. Our study attempted to use MR images in order to develop a deep learning methodology with the use of transfer learning and a convolutional neural network (CNN) ensemble, for the accurate differentiation between the two diseases. An augmented dataset of 210 hips with TOH and 210 hips with AVN was used to finetune three ImageNet-trained CNNs (VGG-16, InceptionResNetV2, and InceptionV3). An ensemble decision was reached in a hard-voting manner by selecting the outcome voted by at least two of the CNNs. Inception-ResNet-V2 achieved the highest AUC (97.62%) similar to the model ensemble, followed by InceptionV3 (AUC of 96.82%) and VGG-16 (AUC 96.03%). Precision for the diagnosis of AVN and recall for the detection of TOH were higher in the model ensemble compared to Inception-ResNet-V2. Ensemble performance was significantly higher than that of an MSK radiologist and a fellow (P < 0.001). Deep learning was highly successful in distinguishing TOH from AVN, with a potential to aid treatment decisions and lead to the avoidance of unnecessary surgery.

Prediction of abdominal aortic aneurysm growth by artificial intelligence taking into account clinical, biologic, morphologic, and biomechanical variables

OBJECTIVES

To develop a prediction model that could risk stratify abdominal aortic aneurysms (AAAs) into high and low growth rate groups, using machine learning algorithms based on variables from different pathophysiological fields.

METHODS

A cohort of 40 patients with small AAAs (maximum diameter 32-53 mm) who had at least an initial and a follow-up CT scan (median follow-up 12 months, range 3-36 months) were included. 29 input variables from clinical, biological, morphometric, and biomechanical pathophysiological aspects extracted for predictive modeling. Collected data were used to build two supervised machine learning models. A gradient boosting (XGboost) and a support vector machines (SVM) algorithm were trained with 60% and tested with 40% of the data to predict which AAA would achieve a growth rate higher than the median of our study cohort. Receiver operating characteristics (ROC) curves and areas under the curve (AUC) were used for the evaluation of the developed algorithms.

RESULTS

XGboost achieved the highest AUC in predicting high compared to low AAA growth rate with an AUC of 81.2% (95% CI from 61.1 to 100%). SVM achieved the second highest performance with an AUC of 68.8% (95% CI from 46.5 to 91%). Based on the best performing algorithm, variable importance was estimated. Diameter-diameter ratio (maximum diameter/neck diameter), Tortuosity from Renal arteries to aortic bifurcation, and maximum thickness of the intraluminal thrombus were found to be the most important factors for model predictions. Other factors were also found to play a significant but less important role.

CONCLUSIONS

A prediction model that can risk stratify AAAs into high and low growth rate groups could be developed by analyzing several factors implicated in the multifactorial pathophysiology of this disease, with the use of machine learning algorithms. Future studies including larger patient cohorts and implementing additional risk markers may aid in the establishment of such methodology during AAA rupture risk estimation.

Research in Musculoskeletal Radiology: Setting Goals and Strategic Directions

The future of musculoskeletal (MSK) radiology is being built on research developments in the field. Over the past decade, MSK imaging research has been dominated by advancements in molecular imaging biomarkers, artificial intelligence, radiomics, and novel high-resolution equipment. Adequate preparation of trainees and specialists will ensure that current and future leaders will be prepared to embrace and critically appraise technological developments, will be up to date on clinical developments, such as the use of artificial tissues, will define research directions, and will actively participate and lead multidisciplinary research. This review presents an overview of the current MSK research landscape and proposes tangible future goals and strategic directions that will fortify the future of MSK radiology.