Opportunistic screening for osteoporosis and osteopenia from CT scans of the abdomen and pelvis using machine learning.

OBJECTIVES: To use multivariable machine learning using the computed tomography (CT) attenuation of each of the bones in the lumbar spine, pelvis, and sacrum, to predict osteoporosis/osteopenia. METHODS: This was a retrospective study of 394 patients aged 50 years or older with CT scans of the abdomen and pelvis and dual-energy x-ray absorptiometry (DXA) scans obtained within 6 months of each other. Volumetric segmentations were performed for each of the bones from L1-L4 vertebrae, pelvis, and sacrum to obtain the mean CT attenuation of each bone. The data was randomly split into training/validation (n = 274, 70%) and test (n = 120, 30%) datasets. The CT attenuation of the L1 vertebrae, univariate logistic regression, least absolute shrinkage and selection operator (LASSO), and support vector machines (SVM) with radial basis function (RBF) were used to predict osteoporosis/osteopenia. The performance of using the CT attenuation at L1 to the univariate logistic regression, LASSO, and SVM models were compared using DeLong's test in the test dataset. RESULTS: All CT attenuation measurements were predictive of osteoporosis/osteopenia (p < 0.001 for all). The SVM model (accuracy = 0.892, AUC = 0.886) outperformed the models using the CT attenuation of threshold of 173.9 Hounsfield units (HU) at L1 (accuracy = 0.725, AUC = 0.739, p = 0.010), the univariate logistic regression model (accuracy = 0.767, AUC = 0.533, p < 0.001) and the LASSO model (accuracy = 0.817, AUC = 0.711, p = 0.007) to predict osteoporosis/osteopenia. CONCLUSION: A SVM model using the CT attenuations of multiple bones within the lumbar spine and pelvis and clinical data has a better ability to predict osteoporosis/osteopenia than using the CT attenuation of L1 or a LASSO model. KEY POINTS: • Multivariable SVM model using the CT attenuation of multiple bones and clinical/demographic data was more predictive than using the CT attenuation at L1 only.

Utilizing machine learning for opportunistic screening for low BMD using CT scans of the cervical spine.

BACKGROUND: Computed Tomography (CT) scans of the cervical spine are often performed to evaluate patients for trauma and degenerative changes of the cervical spine. We hypothesized that the CT attenuation of the cervical vertebrae can be used to identify patients who should be screened for osteoporosis. METHODS: A retrospective study of 253 patients (177 training/validation and 76 test) with unenhanced CT scans of the cervical spine and Dual-energy x-ray Absorbtiometry (DXA) studies within 12 months of each other was performed. Volumetric segmentation of C1-T1, clivus, and first ribs was performed to obtain the CT attenuation of each bone. The correlations of the CT attenuations between the bones and with DXA measurements were evaluated. Univariate receiver operator characteristic (ROC) analyses, and multivariate classifiers (Random Forest (RF), XGBoost, Naïve Bayes (NB), and Support Vector Machines (SVM)) analyzing the CT attenuation of all bones, were utilized to predict patients with osteopenia/osteoporosis and femoral neck bone mineral density (BMD) T-scores <-1. RESULTS: There were positive correlations between the CT attenuation of each bone, and with the DXA measurements. A CT attenuation threshold of 305.2 Hounsfield Units (HU) at C3 had the highest accuracy (0.763, AUC=0.814) to detect femoral neck BMD T-scores ≤-1 and a CT attenuation threshold of 323.6 HU at C3 had the highest accuracy (0.774, AUC=0.843) to detect osteopenia/osteoporosis. The SVM classifier (AUC=0.756) had higher AUC than the RF (AUC=0.692, P=0.224), XGBoost (AUC=0.736; P=0.814), NB (AUC=0.622, P=0.133) and CT threshold of 305.2 HU at C3 (AUC=0.704, P=0.531) classifiers to identify patients with femoral neck BMD T-scores <-1. The SVM classifier (accuracy=0.816) was more accurate than using the CT threshold of 305.2 HU at C3 (accuracy=0.671) (McNemar's χ12=7.55, P=0.006). CONCLUSION: Opportunistic screening for low BMD can be done using cervical spine CT scans. A SVM classifier was more accurate than using the CT threshold of 305.2 HU at C3.

Hepatocellular carcinoma radiation segmentectomy treatment intensification prior to liver transplantation increases rates of complete pathologic necrosis: an explant analysis of 75 tumors.

PURPOSE: To verify the correlation between yttrium-90 glass microsphere radiation segmentectomy treatment intensification of hepatocellular carcinoma (HCC) and complete pathologic necrosis (CPN) at liver transplantation. METHODS: A retrospective, single center, analysis of patients with HCC who received radiation segmentectomy prior to liver transplantation from 2016 to 2021 was performed. The tumor treatment intensification cohort (n = 38) was prescribed radiation segmentectomy as per response recommendations identified in a previously published baseline cohort study (n = 37). Treatment intensification and baseline cohort treatment parameters were compared for rates of CPN. Both cohorts were then combined for an overall analysis of treatment parameter correlation with CPN. RESULTS: Sixty-three patients with a combined 75 tumors were analyzed. Specific activity, dose, and treatment activity were significantly higher in the treatment intensification cohort (all p < 0.01), while particles per cubic centimeter of treated liver were not. CPN was achieved in 76% (n = 29) of tumors in the treatment intensification cohort compared to 49% (n = 18) in the baseline cohort (p = 0.013). The combined cohort CPN rate was 63% (n = 47). ROC analysis showed that specific activity ≥ 327 Bq (AUC 0.75, p < 0.001), dose ≥ 446 Gy (AUC 0.69, p = 0.005), and treatment activity ≥ 2.55 Gbq (AUC 0.71, p = 0.002) were predictive of CPN. Multivariate logistic regression demonstrated that a specific activity ≥ 327 Bq was the sole independent predictor of CPN (p = 0.013). CONCLUSION: Radiation segmentectomy treatment intensification for patients with HCC prior to liver transplantation increases rates of CPN. While dose strongly correlated with pathologic response, specific activity was the most significant independent radiation segmentectomy treatment parameter associated with CPN.

Uterine Artery Embolization for Symptomatic Adenomyosis: Proceedings from a Society of Interventional Radiology Foundation Research Consensus Panel.

Adenomyosis poses an important diagnostic and therapeutic challenge in women's health because of a variety of clinical/imaging presentations and frequent coexistence with other benign gynecologic conditions. In recent years, uterine artery embolization (UAE) for the treatment of adenomyosis has shown encouraging and favorable outcomes and long-term symptom improvement. To expand the current understanding of adenomyosis pathophysiology, imaging diagnostic criteria, and treatment outcomes, the Society of Interventional Radiology Foundation gathered a multidisciplinary Research Consensus Panel with experts from diverse backgrounds. The topics addressed were centered around the following: (i) the clinical presentation and imaging findings to diagnose adenomyosis; (ii) the currently available medical, interventional, and surgical treatment options; and (iii) existing literature for and experiences with UAE in symptomatic disease. The panel acknowledged that before the pursuit of a clinical trial, it would be necessary to first evaluate the imaging criteria for adenomyosis and correlate them with pathology and symptoms to establish a noninvasive imaging classification system. Second priority was given to the development of a quality of life questionnaire to assess patient outcomes following treatment. The third priority was the performance of a prospective clinical trial comparing UAE with medical therapy, which would help establish UAE in the treatment algorithm and societal guidelines for symptomatic adenomyosis.

Radiation Segmentectomy for the Treatment of Solitary Hepatocellular Carcinoma: Outcomes Compared with Those of Surgical Resection.

PURPOSE: To investigate the outcomes of radiation segmentectomy (RS) versus standard-of-care surgical resection (SR). MATERIALS AND METHODS: A multisite, retrospective analysis of treatment-naïve patients who underwent either RS or SR was performed. The inclusion criteria were solitary hepatocellular carcinoma ≤8 cm in size, Eastern Cooperative Oncology Cohort performance status of 0-1, and absence of macrovascular invasion or extrahepatic disease. Target tumor and overall progression, time to progression (TTP), and overall survival rates were assessed. Outcomes were censored for liver transplantation. RESULTS: A total of 123 patients were included (RS, 57; SR, 66). Tumor size, Child-Pugh class, albumin-bilirubin score, platelet count, and fibrosis stage were significantly different between cohorts (P ≤ .01). Major adverse events (AEs), defined as grade ≥3 per the Clavien-Dindo classification, occurred in 0 patients in the RS cohort vs 13 (20%) patients in the SR cohort (P < .001). Target tumor progression occurred in 3 (5%) patients who underwent RS and 5 (8%) patients who underwent SR. Overall progression occurred in 19 (33%) patients who underwent RS and 21 (32%) patients who underwent SR. The median overall TTP was 21.9 and 29.4 months after RS and SR, respectively (95% confidence interval [CI], 15.5-28.2 and 18.5-40.3, respectively; P = .03). Overall TTP subgroup analyses showed no difference between treatment cohorts with fibrosis stages 3-4 (P = .26) and a platelet count of <150 × 109/L (P = .29). The overall progression hazard ratio for RS versus SR was not significant per the multivariate Cox regression analysis (1.16; 95% CI, 0.51-2.63; P = .71). The median overall survival was not reached for either of the cohorts. Propensity scores were calculated but were too dissimilar for analysis. CONCLUSIONS: RS and SR were performed in different patient populations, which limits comparison. RS approached SR outcomes, with a lower incidence of major AEs, in patients who were not eligible for hepatectomy.

Prognosis of breast cancer diagnosed during pregnancy and early postpartum according to immunohistochemical subtype: A matched case-control study.

PURPOSE: Pregnancy-associated breast cancer (PABC) poses a clinical challenge and its prognosis remains controversial. During the pregnancy and postpartum periods, the breast undergoes biological events that may uniquely influence disease behavior and treatment response. This study aimed to assess if a PABC diagnosis influences survival compared to non-PABC. METHODS: A single-center record review was performed to identify PABC patients diagnosed from January 2007 through June 2018. Two controls were matched to each PABC case by stage, immunohistochemical (IHC) subtype, age (± 3) and year of diagnosis (± 2). Disease-free survival (DFS) and overall survival (OS) were estimated with the Kaplan-Meier method and compared with the log-rank test. Multivariate analysis was used to assess the impact of PABC on outcomes. RESULTS: 125 PABC patients (pregnant: 62; postpartum: 63) and 250 controls were included. Median follow-up was 67.7 and 73.4 months, respectively. 4-year DFS was 62% in pregnant vs 78% in controls (p = 0.010), and 63% in postpartum vs 83% in controls (p = 0.034). Subanalysis by IHC subtype revealed a significantly inferior DFS in PABC with hormone receptor-positive/HER2-negative (p = 0.032) and HER2-positive disease (p = 0.005) compared to corresponding non-PABC patients. 4-year OS was similar between case groups and controls. Multivariate analysis supported the independent impact of pregnant and postpartum status on DFS (p < 0.05). CONCLUSION: Patients diagnosed during pregnancy and early postpartum are at high risk of recurrence. Further research is warranted to better characterize PABC tumor biology and enable the identification of novel therapeutic interventions to improve treatment outcomes.

Prioritization of Patients with Abnormal Breast Findings in the Alerta Rosa Navigation Program to Reduce Diagnostic Delays.

INTRODUCTION: In Mexico, there are considerable health system delays in the diagnosis and treatment initiation of women with breast cancer. Alerta Rosa is a navigation program in Nuevo Leon that aims to reduce barriers that impede the timely management of these patients. PATIENTS AND METHODS: Since December 2017, women who registered to receive medical evaluations by Alerta Rosa were stratified based on their clinical characteristics into three priority groups ("Red," "Yellow," and "Green"). According to the category assigned, patients were scheduled imaging studies and medical appointments with breast specialists on a preferential basis. RESULTS: Up until December 2019, 561 patients were scheduled for medical evaluations. Of them, 59% were classified as "Red," 25% "Yellow," and 16% "Green" priority. The median time from stratification to first medical evaluation was 4, 6, and 7 days, respectively (p = .003). Excluding those who had a prior breast cancer diagnosis, 21 patients were diagnosed by Alerta Rosa, with the initial "Red" priority classification demonstrating a sensitivity of 95% (95% confidence interval [CI], 75.1%-99.9%) and specificity of 42% (95% CI, 37.1%-47.1%) for breast cancer. The median time elapsed from initial patient contact to diagnosis and treatment initiation was 16 days and 39 days, respectively. The majority (72%) of patients were diagnosed at an early stage (0-II). CONCLUSION: This patient prioritization system adequately identified women with different probabilities of having breast cancer. Efforts to replicate similar triage systems in resource-constrained settings where screening programs are ineffective could prove to be beneficial in reducing diagnostic intervals and achieving early-stage diagnoses. IMPLICATIONS FOR PRACTICE: Low- and middle-income countries such as Mexico currently lack the infrastructure to achieve effective breast cancer screening and guarantee prompt access to health care when required. To reduce the disease burden in such settings, strategies targeting early detection are urgently needed. Patient navigation programs aid in the reduction of health system intervals and optimize the use of available resources. This article presents the introduction of a triage system based on initial patient concern. Appointment prioritization proved to be successful at reducing health system intervals and achieving early-stage diagnoses by overcoming barriers that impede early access to quality medical care.

Can we screen opportunistically for low BMD using CT scans of the shoulder and artificial intelligence.

OBJECTIVE: To evaluate whether the computed tomography (CT) attenuation of bones seen on shoulder CT scans could be used to predict low bone mineral density (BMD) (osteopenia/osteoporosis), and to compare the performance of two machine learning models to predict low BMD. MATERIALS AND METHODS: 194 patients aged 50 years or greater (69.2 +/- 9.1 years; 170 females) who had unenhanced shoulder CT scans and dual-energy x-ray absorptiometry (DXA) within 1 year of each other between January 1st, 2010 and December 31st, 2021 were evaluated. The CT attenuation of the humerus, glenoid, coracoid, acromion, clavicle, first, second, and third ribs were obtained using 3D-Slicer. Support vector machines (SVM) and k-nearest neighbors (kNN) were used to predict low BMD. DeLong's test was used to compare the area under the curve (AUCs). RESULTS: A CT attenuation of 195.4 Hounsfield Units (HU) of the clavicle had a sensitivity of 0.577, specificity of 0.781 and AUC of 0.701 to predict low BMD. In the test dataset, the SVM had sensitivity of 0.686, specificity of 1.00 and AUC of 0.857; while the kNN model had sensitivity of 0.966, specificity of 0.200, and AUC of 0.583. The SVM was superior to the CT attenuation of the clavicle (P = 0.003), but not better than the kNN model (P = 0.098). CONCLUSION: The CT attenuation of the clavicle was best for predicting low BMD, however, a multivariable SVM was superior for predicting low BMD. ADVANCES IN KNOWLEDGE: SVM utilizing the CT attenuations at many sites was best for predicting low BMD.

Radiation Segmentectomy for the Treatment of Hepatocellular Carcinoma: A Practical Review of Evidence.

Radiation segmentectomy is a versatile, safe, and effective ablative therapy for early-stage hepatocellular carcinoma. Advances in radiation segmentectomy patient selection, procedural technique, and dosimetry have positioned this modality as a curative-intent and guideline-supported treatment for patients with solitary HCC. This review describes key radiation segmentectomy concepts and summarizes the existing literary knowledgebase.

Histologic Findings of Sinusoidal Dilatation and Congestion in Liver Grafts Do Not Correlate with Hepatic Venous Anastomotic Gradients.

PURPOSE: Hepatic venous transplant anastomotic pressure gradient measurement and transjugular liver biopsy are commonly used in clinical decision-making in patients with suspected anastomotic hepatic venous outflow obstruction. This investigation aimed to determine if sinusoidal dilatation and congestion on histology are predictive of hepatic venous anastomotic outflow obstruction, and if it can help select patients for hepatic vein anastomosis stenting. MATERIALS AND METHODS: This is a single-center retrospective study of 166 transjugular liver biopsies in 139 patients obtained concurrently with transplant venous anastomotic pressure gradient measurement. Demographic characteristics, laboratory parameters, procedure and clinical data, and histology of time-zero allograft biopsies were analyzed. RESULTS: No relationship was found between transplant venous anastomotic pressure gradient and sinusoidal dilatation and congestion (P = 0.92). Logistic regression analysis for sinusoidal dilatation and congestion confirmed a significant relationship with reperfusion/preservation injury and/or necrosis of the allograft at time-zero biopsy (OR 6.6 [1.3-33.1], P = 0.02). CONCLUSION: There is no relationship between histologic sinusoidal dilatation and congestion and liver transplant hepatic vein anastomotic gradient. In this study group, sinusoidal dilatation and congestion is a nonspecific histopathologic finding that is not a reliable criterion to select patients for venous anastomosis stenting.

Statistically based nomograms for the minimal needle length required to achieve intra-articular fluoroscopic-guided injections of the shoulder, hip, and knee.

BACKGROUND: Image-guided intra-articular injections are commonly performed to reduce pain in patients with arthritis or other joint-related pathology. Utilizing a needle length that is too short could lead to increased patient discomfort, increased procedural time, and extra-articular injections. OBJECTIVE: To predict the minimum needle length required for fluoroscopic-guided intra-articular injections of the hips, knees, and shoulders based on patient age, gender, height and weight, or body mass index (BMI) and to evaluate whether this varies by gender. STUDY DESIGN: Cross-sectional study. SETTING: Tertiary care academic center. PARTICIPANTS: 600 consecutive patients with available magnetic resonance imaging (MRI) of the hips, knees, and shoulders (100 males and 100 females for each joint). METHODS: The distance from the skin to the joint (glenohumeral, hip and knee) and the thickness of the subcutaneous fat pad (distance from the skin to the muscle) along the injection path were measured. Multivariable linear ridge regression with 10-fold cross-validation was used to predict the distance from the skin to the hip, knee, and glenohumeral joints using age, gender, weight, and height or using age, gender, and BMI. RESULTS: The data show that the subcutaneous fat thickness and the distance from the skin to all joints increase with weight (p < .001) and BMI (p < .001). Subcutaneous fat pads around the anterior shoulder (p < .02) and knee (<.001) are thicker in women than in men. CONCLUSIONS: Patient habitus, in particular weight and BMI, are strong predictors of the thickness of the subcutaneous fat pads and consequently strong predictors of the distance from the skin to the joint. Subcutaneous fat pad thickness around the shoulders and knees varies by gender. Nomograms showing the minimal needle length required to achieve intra-articular injections of the hip, knee and glenohumeral joints are presented.

Detecting whether L1 or other lumbar levels would be excluded from DXA bone mineral density analysis during opportunistic CT screening for osteoporosis using machine learning.

PURPOSE: One or more vertebrae are sometimes excluded from dual-energy X-ray absorptiometry (DXA) analysis if the bone mineral density (BMD) T-score estimates are not consistent with the other lumbar vertebrae BMD T-score estimates. The goal of this study was to build a machine learning framework to identify which vertebrae would be excluded from DXA analysis based on the computed tomography (CT) attenuation of the vertebrae. METHODS: Retrospective review of 995 patients (69.0% female) aged 50 years or greater with CT scans of the abdomen/pelvis and DXA within 1 year of each other. Volumetric semi-automated segmentation of each vertebral body was performed using 3D-Slicer to obtain the CT attenuation of each vertebra. Radiomic features based on the CT attenuation of the lumbar vertebrae were created. The data were randomly split into training/validation (90%) and test datasets (10%). We used two multivariate machine learning models: a support vector machine (SVM) and a neural net (NN) to predict which vertebra(e) were excluded from DXA analysis. RESULTS: L1, L2, L3, and L4 were excluded from DXA in 8.7% (87/995), 9.9% (99/995), 32.3% (321/995), and 42.6% (424/995) patients, respectively. The SVM had a higher area under the curve (AUC = 0.803) than the NN (AUC = 0.589) for predicting whether L1 would be excluded from DXA analysis (P = 0.015) in the test dataset. The SVM was better than the NN for predicting whether L2 (AUC = 0.757 compared to AUC = 0.478), L3 (AUC = 0.699 compared to AUC = 0.555), or L4 (AUC = 0.751 compared to AUC = 0.639) were excluded from DXA analysis. CONCLUSIONS: Machine learning algorithms could be used to identify which lumbar vertebrae would be excluded from DXA analysis and should not be used for opportunistic CT screening analysis. The SVM was better than the NN for identifying which lumbar vertebra should not be used for opportunistic CT screening analysis.

Outcomes of Radiation Segmentectomy for Hepatocellular Carcinoma in Patients with Non-Alcoholic Fatty Liver Disease versus Chronic Viral Hepatitis.

Purpose: To compare the outcomes of radiation segmentectomy for early-stage hepatocellular carcinoma (HCC) in patients with non-alcoholic fatty liver disease (NAFLD) versus hepatitis C virus (HCV). Materials and Methods: A retrospective analysis of consecutive patients with NAFLD- or HCV-related HCC treated with radiation segmentectomy from 01/2017-06/2022 was performed. Eligibility criteria included solitary tumor ≤8 cm or up to 3 HCC ≤3 cm, ECOG 0-1, and absence of vascular invasion or extrahepatic spread. Imaging best response was assessed per modified Response Evaluation Criteria in Solid Tumors. Target tumor and overall progression, time-to-progression (TTP), and overall survival (OS) were calculated. All outcomes were censored for liver transplantation (LT). Complete pathologic response (CPN) was assessed in patients who underwent LT. Results: Of 142 patients included (NAFLD: 61; HCV: 81), most had cirrhosis (NAFLD: 87%; HCV: 86%) and small tumors (median size NAFLD: 2.3 cm; HCV: 2.5 cm). Patients with NAFLD had higher BMI (p<0.001) and worse ALBI scores (p=0.003). Patients with HCV were younger (p<0.001) and had higher AFP levels (p=0.034). Median radiation dose (NAFLD: 508 Gy; HCV: 452 Gy) and specific activity (NAFLD: 700 Bq; HCV: 698 Bq) were similar between cohorts. Objective response was 100% and 97% in the NAFLD and HCV cohorts, respectively. Target tumor progression occurred in 1 (2%) NAFLD and 8 (10%) HCV patients. Target tumor TTP was not met for either cohort. Overall progression occurred in 23 (38%) NAFLD and 39 (48%) HCV patients. Overall TTP was 17.4 months (95% CI 13.5-22.2) in NAFLD and 13.5 months (95% CI 0.4-26.6) in HCV patients (p=0.86). LT was performed in 27 (44%) NAFLD and 33 (41%) HCV patients, with a CPN rate of 63% and 54%, respectively. OS was not met in the NAFLD cohort and was 53.9 months (95% CI 32.1-75.7) in the HCV cohort (p=0.15). Conclusion: Although NAFLD and HCV are associated with different mechanisms of liver injury, patients with early-stage HCC treated with radiation segmentectomy achieve comparable outcomes.

Effect of receiving a customizable brochure on breast cancer patients' knowledge about their diagnosis and treatment: A randomized clinical trial.

BACKGROUND: Patients' lack of knowledge about their own disease may function as a barrier to shared decision-making and well-being. This study aimed to evaluate the impact of written educational materials on breast cancer patients. METHODS: This multicenter, parallel, unblinded, randomized trial included Latin American women aged ≥18 years with a recent breast cancer diagnosis yet to start systemic therapy. Participants underwent randomization in a 1:1 ratio to receive a customizable or standard educational brochure. The primary objective was accurate identification of molecular subtype. Secondary objectives included identification of clinical stage, treatment options, participation in decision-making, perceived quality of information received, and illness uncertainty. Follow-up occurred at 7-21 and 30-51 days post-randomization. CLINICALTRIALS: gov identifier: NCT05798312. RESULTS: One hundred sixty-five breast cancer patients with a median age of 53 years and 61 days from diagnosis were included (customizable: 82; standard: 83). At first available assessment, 52%, 48%, and 30% identified their molecular subtype, disease stage, and guideline-endorsed systemic treatment strategy, respectively. Accurate molecular subtype and stage identification were similar between groups. Per multivariate analysis, customizable brochure recipients were more likely to identify their guideline-recommended treatment modalities (OR: 4.20,p = 0.001). There were no differences between groups in the perceived quality of information received or illness uncertainty. Customizable brochure recipients reported increased participation in decision-making (p = 0.042). CONCLUSIONS: Over one third of recently diagnosed breast cancer patients are incognizant of their disease characteristics and treatment options. This study demonstrates a need to improve patient education and shows that customizable educational materials increase patients' understanding of recommended systemic therapies according to individual breast cancer characteristics.

Machine Learning for Opportunistic Screening for Osteoporosis from CT Scans of the Wrist and Forearm.

Background: We investigated whether opportunistic screening for osteoporosis can be done from computed tomography (CT) scans of the wrist/forearm using machine learning. Methods: A retrospective study of 196 patients aged 50 years or greater who underwent CT scans of the wrist/forearm and dual-energy X-ray absorptiometry (DEXA) scans within 12 months of each other was performed. Volumetric segmentation of the forearm, carpal, and metacarpal bones was performed to obtain the mean CT attenuation of each bone. The correlations of the CT attenuations of each of the wrist/forearm bones and their correlations to the DEXA measurements were calculated. The study was divided into training/validation (n = 96) and test (n = 100) datasets. The performance of multivariable support vector machines (SVMs) was evaluated in the test dataset and compared to the CT attenuation of the distal third of the radial shaft (radius 33%). Results: There were positive correlations between each of the CT attenuations of the wrist/forearm bones, and with DEXA measurements. A threshold hamate CT attenuation of 170.2 Hounsfield units had a sensitivity of 69.2% and a specificity of 77.1% for identifying patients with osteoporosis. The radial-basis-function (RBF) kernel SVM (AUC = 0.818) was the best for predicting osteoporosis with a higher AUC than other models and better than the radius 33% (AUC = 0.576) (p = 0.020). Conclusions: Opportunistic screening for osteoporosis could be performed using CT scans of the wrist/forearm. Multivariable machine learning techniques, such as SVM with RBF kernels, that use data from multiple bones were more accurate than using the CT attenuation of a single bone.

Support vector machines are superior to principal components analysis for selecting the optimal bones' CT attenuations for opportunistic screening for osteoporosis using CT scans of the foot or ankle.

Objectives: To use the computed tomography (CT) attenuation of the foot and ankle bones for opportunistic screening for osteoporosis. Methods: Retrospective study of 163 consecutive patients from a tertiary care academic center who underwent CT scans of the foot or ankle and dual-energy X-ray absorptiometry (DXA) within 1 year of each other. Volumetric segmentation of each bone of the foot and ankle was done in 3D Slicer to obtain the mean CT attenuation. Pearson's correlations were used to correlate the CT attenuations with each other and with DXA measurements. Support vector machines (SVM) with various kernels and principal components analysis (PCA) were used to predict osteoporosis and osteopenia/osteoporosis in training/validation and test datasets. Results: CT attenuation measurements at the talus, calcaneus, navicular, cuboid, and cuneiforms were correlated with each other and positively correlated with BMD T-scores at the L1-4 lumbar spine, hip, and femoral neck; however, there was no significant correlation with the L1-4 trabecular bone scores. A CT attenuation threshold of 143.2 Hounsfield units (HU) of the calcaneus was best for detection of osteoporosis in the training/validation dataset. SVMs with radial basis function (RBF) kernels were significantly better than the PCA model and the calcaneus for predicting osteoporosis in the test dataset. Conclusions: Opportunistic screening for osteoporosis is possible using the CT attenuation of the foot and ankle bones. SVMs with RBF using all bones is more accurate than the CT attenuation of the calcaneus.

Machine learning for the prediction of osteopenia/osteoporosis using the CT attenuation of multiple osseous sites from chest CT.

PURPOSE: To use machine learning and the CT attenuation of all bones visible on chest CT scans to predict osteopenia/osteoporosis. METHOD: We retrospectively evaluated 364 patients with CT scans of the chest, and Dual-energy X-ray absorptiometry (DXA) scans within 6 months of each other. Studies were performed between 01/01/2015 and 08/01/2021. Volumetric segmentation of the ribs, thoracic vertebrae, sternum, and clavicle was performed using 3D Slicer to obtain the mean CT attenuation of each bone. The study sample was randomly split into training/validation (80 %, n = 291 patients) and test (20 %, n = 73 patients) datasets. Univariate analyses were used to identify the optimal CT attenuation thresholds to diagnose osteopenia/osteoporosis. We used penalized multivariable logistic regression models including Least Absolute Shrinkage and Selection Operator (LASSO), Elastic Net, and Ridge regression, and Support Vector Machines (SVM) with radial basis functions (RBF) to predict osteopenia/osteoporosis and compared these results to the CT attenuation threshold at T12. RESULTS: There were positive correlations between the CT attenuation between all bones (r > 0.6, P < 0.001 for all). There were positive correlations between CT attenuation of the bones and the L1-L4 BMD T-score, total hip T-score, and femoral neck T-scores (r > 0.4, P < 0.001 for all). A CT attenuation threshold of 170.2 Hounsfield units (HU) at T12 had an AUC of 0.702, while a threshold of 192.1 HU at T4 had an AUC of 0.757. The SVM with RBF had the highest AUC (AUC = 0.864) and was better than the LASSO (P = 0.011), Elastic Net (P = 0.011), Ridge regression (P = 0.011) but was not better than using the CT attenuation at T12 (P = 0.060). CONCLUSIONS: The CT attenuation of the ribs, thoracic vertebra, sternum, and clavicle can be used individually and collectively to predict BMD and to predict osteopenia/osteoporosis.

Correction: Sebro, R.; De la Garza-Ramos, C. Machine Learning for Opportunistic Screening for Osteoporosis from CT Scans of the Wrist and Forearm. Diagnostics 2022, 12, 691.

In the original publication [...].

Nononcologic Embolization.

The scope of conditions managed by embolization, which was initially used for the treatment of hemorrhage and vascular malformations, is constantly expanding. Apart from oncologic indications, embolization is used to treat a wide range of benign pathology, including uterine fibroids and benign prostatic hyperplasia. While various particulate embolic agents are successfully used for benign embolization, there is growing evidence that unique properties of these may result in different outcomes. This article reviews available evidence comparing various particles used for uterine fibroid embolization and prostate artery embolization. In addition, we provide an overview of periprocedural pharmacology and protocols facilitating same-day discharge for these interventions.