Diagnostic value of machine learning-based computed tomography texture analysis for differentiating multiple myeloma from osteolytic metastatic bone lesions in the peripheral skeleton.

OBJECTIVES: To report the diagnostic performance of machine learning-based CT texture analysis for differentiating multiple myeloma from osteolytic metastatic bone lesions in the peripheral skeleton. METHODS: We retrospectively evaluated 172 patients with multiple myeloma (n = 70) and osteolytic metastatic bone lesions (n = 102) in the peripheral skeleton. Two radiologists individually used two-dimensional manual segmentation to extract texture features from non-contrast CT. In total, 762 radiomic features were extracted. Dimension reduction was performed in three stages: inter-observer agreement analysis, collinearity analysis, and feature selection. Data were randomly divided into training (n = 120) and test (n = 52) groups. Eight machine learning algorithms were used for model development. The primary performance metrics were the area under the receiver operating characteristic curve and accuracy. RESULTS: In total, 476 of the 762 texture features demonstrated excellent interobserver agreement. The number of features was reduced to 22 after excluding those with strong collinearity. Of these features, six were included in the machine learning algorithms using the wrapper-based classifier-specific technique. When all eight machine learning algorithms were considered for differentiating multiple myeloma from osteolytic metastatic bone lesions in the peripheral skeleton, the area under the receiver operating characteristic curve and accuracy were 0.776-0.932 and 78.8-92.3%, respectively. The k-nearest neighbors model performed the best, with the area under the receiver operating characteristic curve and accuracy values of 0.902 and 92.3%, respectively. CONCLUSION: Machine learning-based CT texture analysis is a promising method for discriminating multiple myeloma from osteolytic metastatic bone lesions.

Sonoelastography findings of myofibroblastoma.

Myofibrolastoma of the breast (MFB) is a rare and benign stromal tumor. Although MFB is a benign entity, in radiologic evaluation, there is no specific diagnostic feature. Conventional breast imaging findings have been published before. Sonoelastography has been used as an imaging method that allows us to evaluate tissue stiffness in vivo and increases the specificity of B-mode ultrasonography in the discrimination of benign-malignant lesions. In this case report, it was shown that MFB is of high stiffness in shear wave elastography (SWE) evaluation. SWE findings of MFB, which is a benign lesion, have been described for the first time in the literature.

Evaluation of hereditary/familial breast cancer patients with multigene targeted next generation sequencing panel and MLPA analysis in Turkey.

Breast cancer, a worldwide leading cause of cancer in women, may occur in familial cases. Germline mutations in BRCA1/2 genes are responsible for 15% of the familial cases. With the power of next generation sequencing (NGS) analysis, it is possible to analyze genes related to hereditary susceptibility to breast cancer and investigate the genetic etiology more thoroughly. In this study, we investigated 30 genes identified frequent pathogenic alleles in Turkish population. The study includes 495 unrelated individuals diagnosed with breast cancer who are selected for genetic testing according to NCCN criteria for hereditary breast cancer. All patients were analyzed by NGS for BRCA1/2 genes. Deletion/duplication investigation by Multiplex ligation-dependent probe amplification (MLPA) and massive sequencing of 30 breast cancer-related genes (Oncorisk Gene Panel) were performed in a stepwise manner. BRCA1/2 variants are the most frequent pathogenic variants which are found in 45 of 495 (9.1%) patients. Four previously unreported, novel, pathogenic variants of BRCA2 gene are identified. In four cases, exonic deletions of BRCA1/2 genes are determined and there is no duplication of these genes. NGS panel investigation involving other moderate-high risk genes contributed genetic diagnosis in an extra 39 out of 419 (9.3%) cases. Our study presents the cost effectiveness of the gene panel approach. We suggest that gene panels should be the first-tier genetic testing for hereditary breast cancer and MLPA analysis of BRCA1/2 genes should be investigated as a complementary method of NGS analysis.

Giant gossypiboma presenting as a pelvic mass.

Gossypiboma is a foreign object, such as a mass of cotton matrix or a sponge, that is left behind in a body cavity during surgery. It is uncommon, mostly asymptomatic, and hard to diagnose. It may be incorrectly diagnosed preoperatively, which can lead to unnecessary invasive diagnostic procedures and operations. It should be included in the differential diagnosis of soft-tissue masses detected in patients with a history of a prior operation. We present a case of 36-year-old female who referred to emergency room with severe abdominal pain and distension. Imaging revealed a giant intra-abdominal mass resembling a soft tissue tumor, but revealed to be a giant gossypiboma caused by a sponge that was forgotten during previous ectopic pregnancy surgery. This case differs from others with the absence of findings supporting gossypiboma such as calcification or trapped gas bubbles and emphasizes the importance of this potentially life-threatening complication of surgery.

Increased prevalence of pelvic venous congestion sign on sacroiliac MRI in women with clinically suspected sacroiliitis.

OBJECTIVE: To compare the prevalence of pelvic venous congestion (PVC) sign on sacroiliac and hip MRI in women of reproductive age as a possible cause of pain mimicking sacroiliitis. METHODS: This retrospective study included 727 MRI examinations (401 sacroiliac joint MRI and 326 hip joint MRI) that were performed between January 2010 and December 2017. Images were evaluated for the presence of sacroiliitis, presence of PVC sign and presence of other incidental findings of musculoskeletal and genitourinary disorders. After removing patients with other musculoskeletal and genitourinary disorders that may cause pain (n=188), remaining 539 (322 sacroiliac and 217 hip), MRI examinations were re-analyzed for the presence of PVC sign. RESULTS: Four hundred one patients with sacroiliac MRI examination had 120 (29.9%) PVC sign and 326 patients with hip MRI examinations had 54 (16.6%) PVC sign (p<0.001). After removing patients with other musculoskeletal and genitourinary disorders that may cause pain, 322 patients with sacroiliac MRI had 102 (31.7%) PVC sign and 217 patients with hip MRI examinations had 38 (17.5%) PVC sign (p<0.001). No significant difference was found between patients with acute sacroiliitis and patients without acute sacroiliitis concerning PVC prevalence (p>0.05). There were also no significant differences between other comparable incidental findings. CONCLUSION: Significantly increased PVC prevalence in sacroiliac MRI exams may be attributable to PCS simulating clinical sacroiliitis.

Discrimination of oncocytoma and chromophobe renal cell carcinoma using MRI.

PURPOSE: We aimed to evaluate magnetic resonance imaging (MRI) features, including signal intensities, enhancement patterns and T2 signal intensity ratios to differentiate oncocytoma from chromophobe renal cell carcinoma (RCC). METHODS: This retrospective study included 17 patients with oncocytoma and 33 patients with chromophobe RCC who underwent dynamic MRI. Two radiologists independently reviewed images blinded to pathology. Morphologic characteristics, T1 and T2 signal intensities were reviewed. T2 signal intensities, wash-in, wash-out values, T2 signal intensity ratios were calculated. Sensitivity and specificity analyses were performed. RESULTS: Mean ages of patients with oncocytoma and chromophobe RCC were 61.0±11.6 and 58.5±14.0 years, respectively. Mean tumor size was 60.6±47.3 mm for oncocytoma, 61.7±45.9 mm for chromophobe RCC. Qualitative imaging findings in conventional MRI have no distinctive feature in discrimination of two tumors. Regarding signal intensity ratios, oncocytomas were higher than chromophobe RCCs. Renal oncocytomas showed higher signal intensity ratios and wash-in values than chromophobe RCCs in all phases. Fast spin-echo T2 signal intensities were higher in oncocytomas than chromophobe RCCs. CONCLUSION: Signal intensity ratios, fast spin-echo T2 signal intensities and wash-in values constitute diagnostic parameters for discriminating between oncoytomas and chromophobes. In the excretory phase of dynamic enhanced images, oncocytomas have higher signal intensity ratio than chromophobe RCC and high wash-in values strongly imply a diagnosis of renal oncocytoma.

How high is the inter-observer reproducibility in the LIRADS reporting system?

PURPOSE: To investigate the reproducibility of LIRADS v2014 and contribute to its widespread use in clinical practice. MATERIAL AND METHODS: This retrospective, single-centre study was conducted between January 2010 and October 2015. A total of 132 patients who had dynamic magnetic resonance imaging (MRI)/computed tomography (CT) images in the Picture Archiving and Communication Systems (PACS) with liver nodule were included in the study, 37 of whom had histopathology results. Five radiologists who participated in the study, interpreted liver nodules independently on different PACS stations according to the LIRADS reporting system and its main parameters. RESULTS: We determined that level of inter-observer agreement in the LR-1, LR-5, and LR-5V categories was higher than in the LR-2, LR-3, and LR-4 categories (κ = 0.522, 0.442, and 0.600 in the LR-1, LR-5, and LR-5V categories, respectively; κ = 0.082, 0.298, and 0.143 in the LR-2, LR-3, and LR-4 categories, respectively). The parameter that we observed to have the highest level of inter-observer agreement was venous thrombus (κ = 0.600). CONCLUSIONS: Our study showed that LIRADS achieves an acceptable inter-observer reproducibility in terms of clinical practice although it is insufficient at intermediate risk levels. We think that the prevalence of its use will be further increased with training related to the subject and the assignment of numerical values that express the probability of malignancy for each category and including the ancillary features in the algorithm according to clearer rules.

Utility of chest CT after a chest X-ray in patients presenting to the ED with non-traumatic thoracic emergencies.

BACKGROUND: Chest radiography is the initial choice for thoracic imaging. However, the wide availability of computed tomography (CT) has led to a substantial increase in its use in the emergency department (ED). We evaluated the utility of chest CT after a chest X-ray in patients presenting to the ED with non-traumatic thoracic emergencies, and determined if the diagnosis and management decision changed after CT. METHODS: The study enrolled 500 consecutive patients with both chest X-rays and CT who presented to the ED with non-traumatic complaints. Chest X-rays and CT images obtained within 12h before any definitive treatment were randomly evaluated in consensus by two radiologists blinded to the clinical information. RESULTS: The chest X-ray and CT image findings were concordant in 49.2% of the 500 patients and this concordance was negatively correlated with patient age. Leading diagnosis and management decisions based on the chest radiograph changed after CT in 35.4% of the study group and this finding was also correlated with age. In 55% of 205 patients, pneumonic infiltrations were undiagnosed with radiography. Pulmonary edema was the most specific (93.3%) and sensitive (85.4%) radiography finding. Posteroanterior chest radiographs taken in the upright position had higher concordance with CT than anteroposterior (AP) radiographs taken in the supine position. CONCLUSIONS: Chest CT may be an appropriate imaging choice in patients presenting to the ED for non-traumatic reasons, particularly for elderly patients and when the radiograph is taken with the AP technique in a supine position.