Multimodal modeling with low-dose CT and clinical information for diagnostic artificial intelligence on mediastinal tumors: a preliminary study.

BACKGROUND: Diagnosing mediastinal tumours, including incidental lesions, using low-dose CT (LDCT) performed for lung cancer screening, is challenging. It often requires additional invasive and costly tests for proper characterisation and surgical planning. This indicates the need for a more efficient and patient-centred approach, suggesting a gap in the existing diagnostic methods and the potential for artificial intelligence technologies to address this gap. This study aimed to create a multimodal hybrid transformer model using the Vision Transformer that leverages LDCT features and clinical data to improve surgical decision-making for patients with incidentally detected mediastinal tumours. METHODS: This retrospective study analysed patients with mediastinal tumours between 2010 and 2021. Patients eligible for surgery (n=30) were considered 'positive,' whereas those without tumour enlargement (n=32) were considered 'negative.' We developed a hybrid model combining a convolutional neural network with a transformer to integrate imaging and clinical data. The dataset was split in a 5:3:2 ratio for training, validation and testing. The model's efficacy was evaluated using a receiver operating characteristic (ROC) analysis across 25 iterations of random assignments and compared against conventional radiomics models and models excluding clinical data. RESULTS: The multimodal hybrid model demonstrated a mean area under the curve (AUC) of 0.90, significantly outperforming the non-clinical data model (AUC=0.86, p=0.04) and radiomics models (random forest AUC=0.81, p=0.008; logistic regression AUC=0.77, p=0.004). CONCLUSION: Integrating clinical and LDCT data using a hybrid transformer model can improve surgical decision-making for mediastinal tumours, showing superiority over models lacking clinical data integration.

Review of clinical and diagnostic imaging of the thymus: from age-related changes to thymic tumors and everything in between.

The thymus, a primary lymphoid organ of the immune system, undergoes several changes due to a variety of reasons, ranging from aging to pathological conditions. These changes can make distinguishing between benign and neoplastic changes in the thymus challenging, thereby complicating the histopathological diagnoses of thymic tumors. Moreover, most patients with thymic tumors are asymptomatic at the time of diagnosis. Therefore, imaging plays an extremely important role in the evaluation of thymic lesions. In this review, we introduced the imaging characteristics of the thymus, ranging from benign findings, such as normal maturation and benign lesions, to neoplasms.

Intravascular large B-cell lymphoma appearance on dual-energy computed tomography: a case report.

BACKGROUND: Intravascular large B-cell lymphoma (IVLBCL) is the proliferation of neoplastic B lymphocytes in the vascular space. Since conventional computed tomography (CT) shows nonspecific findings, differentiation between IVLBCL and other lung diseases, such as diffuse interstitial lung disease, is difficult. CASE PRESENTATION: A 73-year-old man presented with dyspnea and hypoxemia. Laboratory findings showed an increased lactate dehydrogenase level of 1690 U/L (normal: 130-235 U/L) and soluble interleukin-2 receptor level of 1140 U/mL (normal: 157-474U/mL). Dual-energy CT iodine mapping showed a significant symmetrical decrease in iodine distribution in the upper lungs, suggesting an unusual distribution of pulmonary hypoperfusion. Therefore, IVLBCL was suspected. A random skin biopsy confirmed the diagnosis of IVLBCL. Due to the severity of the disease, lung biopsy was averted. After admission to the hospital, high-dose methotrexate was administered for central nervous system involvement, due to findings of suspected intracranial infiltration on a brain magnetic resonance imaging and elevated cell counts on lumbar puncture. Subsequently, oxygen demand improved, and rituximab along with cyclophosphamide, doxorubicin, vincristine, and prednisone was added to the patient's regime. Eventually, oxygen administration was terminated, the patient's general condition improved, and the patient was discharged after 47 days of hospitalization. CONCLUSIONS: Since the diagnosis of IVLBCL depends on whether it is possible to suspect IVLBCL, the finding of decreased iodine perfusion demonstrated on dual-energy CT is considered important information for diagnosis. An immediate diagnosis of IVLBCL is needed to avoid rapid disease progression and introduce early treatment for a favorable prognosis. In this case, unique pulmonary hypoperfusion demonstrated by dual-energy CT promoted early diagnosis of IVLBCL.

Ex-vivo 1.5T MR Imaging versus CT in Estimating the Size of the Pathologically Invasive Component of Lung Adenocarcinoma Spectrum Lesions.

PURPOSE: The purpose of this study was to investigate whether ex-vivo MRI enables accurate estimation of the invasive component of lung adenocarcinoma. METHODS: We retrospectively reviewed 32 patients with lung adenocarcinoma who underwent lung lobectomy. The specimens underwent MRI at 1.5T. The boundary between the lesion and the normal lung was evaluated on a 5-point scale in each three MRI sequences, and a one-way analysis of variance and post-hoc tests were performed. The invasive component size was measured histopathologically. The maximum diameter of each solid component measured on CT and MR T1-weighted (T1W) images and the maximum size obtained from histopathologic images were compared using the Wilcoxon signed-rank test. Inter-reader agreement was evaluated using intraclass correlation coefficients (ICC). RESULTS: T1W images were determined to be optimal for the delineation of the lesions (P < 0.001). The histopathologic invasive area corresponded to the area where the T1W ex-vivo MR image showed a high signal intensity that was almost equal to the intravascular blood signal. The maximum diameter of the solid component on CT was overestimated compared with the maximum invasive size on histopathology (mean, 153%; P < 0.05), while that on MRI was evaluated mostly accurately without overestimation (mean, 108%; P = 0.48). The interobserver reliability of the measurements using CT and MRI was good (ICC = 0.71 on CT, 0.74 on MRI). CONCLUSION: Ex-vivo MRI was more accurate than conventional CT in delineating the invasive component of lung adenocarcinoma.

Visual classification of three computed tomography lung patterns to predict prognosis of COVID-19: a retrospective study.

BACKGROUND: Quantitative evaluation of radiographic images has been developed and suggested for the diagnosis of coronavirus disease 2019 (COVID-19). However, there are limited opportunities to use these image-based diagnostic indices in clinical practice. Our aim in this study was to evaluate the utility of a novel visually-based classification of pulmonary findings from computed tomography (CT) images of COVID-19 patients with the following three patterns defined: peripheral, multifocal, and diffuse findings of pneumonia. We also evaluated the prognostic value of this classification to predict the severity of COVID-19. METHODS: This was a single-center retrospective cohort study of patients hospitalized with COVID-19 between January 1st and September 30th, 2020, who presented with suspicious findings on CT lung images at admission (n = 69). We compared the association between the three predefined patterns (peripheral, multifocal, and diffuse), admission to the intensive care unit, tracheal intubation, and death. We tested quantitative CT analysis as an outcome predictor for COVID-19. Quantitative CT analysis was performed using a semi-automated method (Thoracic Volume Computer-Assisted Reading software, GE Health care, United States). Lungs were divided by Hounsfield unit intervals. Compromised lung (%CL) volume was the sum of poorly and non-aerated volumes (- 500, 100 HU). We collected patient clinical data, including demographic and clinical variables at the time of admission. RESULTS: Patients with a diffuse pattern were intubated more frequently and for a longer duration than patients with a peripheral or multifocal pattern. The following clinical variables were significantly different between the diffuse pattern and peripheral and multifocal groups: body temperature (p = 0.04), lymphocyte count (p = 0.01), neutrophil count (p = 0.02), c-reactive protein (p < 0.01), lactate dehydrogenase (p < 0.01), Krebs von den Lungen-6 antigen (p < 0.01), D-dimer (p < 0.01), and steroid (p = 0.01) and favipiravir (p = 0.03) administration. CONCLUSIONS: Our simple visual assessment of CT images can predict the severity of illness, a resulting decrease in respiratory function, and the need for supplemental respiratory ventilation among patients with COVID-19.

Incidental mediastinal masses detected at low-dose CT screening: prevalence and radiological characteristics.

PURPOSE: The primary and secondary aims were to investigate the prevalence of incidental mediastinal masses on low-dose chest CT examinations during health check-ups, and to review the radiological characteristics of prevascular mediastinal masses, respectively. MATERIALS AND METHODS: This retrospective study included 38,861 participants (mean age: 57.1 years; range: 21-99 years; men: 51.3%; never-smokers: 57.4%) who underwent low-dose chest CT examinations between January 2011 and December 2016. All images with incidental mediastinal masses were reviewed, and prevascular mediastinal masses were assessed for qualitative and quantitative imaging characteristics by two radiologists. Univariate and multivariate analyses were performed in clinical and CT features between some combinations of participants. RESULTS: Overall, 653 participants (1.68%, 653 of 38,861) had incidental mediastinal masses; 578 in prevascular mediastinum, including 93 intrathymic cysts and 24 thymic epithelial tumors. Presence of mediastinal mass was not significantly associated with sex (p = 0.089) and smoking history (p = 0.098) but with age (p < 0.001). Significant differences were found between intrathymic cysts and thymic epithelial tumors in terms of shapes (p = 0.049), contours (p = 0.018), and CT values (p = 0.012). CONCLUSION: The prevalence of asymptomatic mediastinal masses on low-dose chest CT was 1.68%. CT values, shapes, and contours may effectively distinguish intrathymic cysts from thymic epithelial tumors.

MR Imaging of Mushroom-like Skin Adnexal Tumors in the Scalp: A Report of Two Cases.

Few studies had been published regarding imaging findings of skin adnexal tumors. We experienced two giant cases of them with a characteristic mushroom-like growth pattern. MRI showed a circumscribed mushroom-like shaped mass extruding from the subcutaneous tissue with microcystic lesions. Although differentiation between benignancy and malignancy may be difficult by radiological examinations, MRI may be helpful to identify its origin and differentiate soft tissue tumors with skin adnexal tumors in having these imaging findings.

Pediatric oncologic emergencies: Clinical and imaging review for pediatricians.

Children with cancer are at increased risk of life-threatening emergencies, either from the cancer itself or related to the cancer treatment. These conditions need to be assessed and treated as early as possible to minimize morbidity and mortality. Cardiothoracic emergencies encompass a variety of pathologies, including pericardial effusion and cardiac tamponade, massive hemoptysis, superior vena cava syndrome, pulmonary embolism, and pneumonia. Abdominal emergencies include bowel obstruction, intussusception, perforation, tumor rupture, intestinal graft-versus-host disease, acute pancreatitis, neutropenic colitis, and obstructive uropathy. Radiology plays a vital role in the diagnosis of these emergencies. We here review the clinical features and imaging in pediatric patients with oncologic emergencies, including a review of recently published studies. Key radiological images are presented to highlight the radiological approach to diagnosis. Pediatricians, pediatric surgeons, and pediatric radiologists need to work together to arrive at the correct diagnosis and to ensure prompt and appropriate treatment strategies.

Value of a Computer-aided Detection System Based on Chest Tomosynthesis Imaging for the Detection of Pulmonary Nodules.

Purpose To assess the value of a computer-aided detection (CAD) system for the detection of pulmonary nodules on chest tomosynthesis images. Materials and Methods Fifty patients with and 50 without pulmonary nodules underwent both chest tomosynthesis and multidetector computed tomography (CT) on the same day. Fifteen observers (five interns and residents, five chest radiologists, and five abdominal radiologists) independently evaluated tomosynthesis images of 100 patients for the presence of pulmonary nodules in a blinded and randomized manner, first without CAD, then with the inclusion of CAD marks. Multidetector CT images served as the reference standard. Free-response receiver operating characteristic analysis was used for the statistical analysis. Results The pooled diagnostic performance of 15 observers was significantly better with CAD than without CAD (figure of merit [FOM], 0.74 vs 0.71, respectively; P = .02). The average true-positive fraction and false-positive rate per all cases with CAD were 0.56 and 0.26, respectively, whereas those without CAD were 0.47 and 0.20, respectively. Subanalysis showed that the diagnostic performance of interns and residents was significantly better with CAD than without CAD (FOM, 0.70 vs 0.62, respectively; P = .001), whereas for chest radiologists and abdominal radiologists, the FOM with CAD values were greater but not significantly: 0.80 versus 0.78 (P = .38) and 0.74 versus 0.73 (P = .65), respectively. Conclusion CAD significantly improved diagnostic performance in the detection of pulmonary nodules on chest tomosynthesis images for interns and residents, but provided minimal benefit for chest radiologists and abdominal radiologists. © RSNA, 2017 Online supplemental material is available for this article.

Imaging characteristics of subcutaneous amyloid deposits in diabetic patients: the "insulin ball".

OBJECTIVE: The purpose of this study was to describe the imaging characteristics of subcutaneous amyloid deposits occurring at sites of insulin injection, commonly known as "insulin balls," in diabetic patients on ultrasound, CT, and MRI with pathologic correlation. MATERIALS AND METHODS: We retrospectively reviewed the radiographic findings of 14 lesions in 9 patients diagnosed with subcutaneous amyloid deposits at our institution between 2005-2015. Three board-certified radiologists analyzed the following: (1) the shape, size, margin, morphologic characteristics, and blood flow on US using the color Doppler signal, (2) shape, size, margin, attenuation, and presence or absence of contrast enhancement on CT, and (3) shape, size, margin, signal intensity, and presence or absence of contrast enhancement on MRI. RESULTS: All lesions showed ill-defined hypovascular subcutaneous nodules with irregular margins. The median diameter of lesions was 50.4 mm on US, 46.8 mm on CT, and 51.4 mm on MRI. The internal echogenicity of subcutaneous amyloid deposits was hypoechoic and heterogeneous on US. All lesions showed isodensity compared to muscle with irregular margins and minimal contrast enhancement on CT. Both T1- and T2-weighted MR images showed low signal intensity compared with subcutaneous fat. Normal diffusion and minimal contrast enhancement were seen. CONCLUSIONS: Subcutaneous amyloid deposits which cause insulin resistance are typically ill-defined and heterogeneous hypovascular subcutaneous nodules with irregular margins on imaging that correspond to insulin injection sites. It is also characteristic that T2WI shows low intensity compared with fat on MRI, reflective of the amyloid content.

Marathon-induced pulmonary edema of a patient with transient dyspnea.

We report a case of a 31-year-old healthy man with marathon-induced pulmonary edema. Chest radiograph revealed pulmonary edema without cardiomegaly. Contrast-enhanced chest computed tomography (CT) revealed transient pulmonary edema without filling-defect in pulmonary arteries. As marathon running increases in popularity, radiologists and emergency physicians should be familiar with diagnosis of this entity on chest radiograph, avoiding unnecessary CT examination without additional clinical information.

Radiology of Fractures in Intoxicated Emergency Department Patients: Locations, Mechanisms, Presentation, and Initial Interpretation Accuracy.

The purpose of this study was to investigate the relationship of alcohol intoxication to time-to-presentation following injury, fracture type, mechanism of injury leading to fracture, and initial diagnostic radiology interpretation performance of emergency physicians versus diagnostic radiologists in patients who present to the emergency department (ED) and are subsequently diagnosed with fracture. Medical records of 1286 patients who presented to the ED and were diagnosed with fracture who also underwent plain film or computed tomography (CT) imaging were retrospectively reviewed. The subjects were divided into intoxicated and sober groups. Patient characteristics, injury-to-presentation time, fracture location, and discrepancies between initial clinical and radiological evaluations were compared. Of 1286 subjects, 181 patients were included in the intoxicated group. Only intoxicated patients presented with head/neck fractures more than 24 hours after injury. The intoxicated group showed a higher rate of head/neck fractures (skull 23.2% vs 5.8%, face and orbit 30.4% vs 9.5%; P < 0.001) and a lower rate of extremity injuries. The rate of nondiagnosis of fractures by emergency physicians later identified by radiologists was the same in both groups (7.7% vs 7.7%, P = 0.984). While the same proportion of intoxicated patients presented more than 24 hours following injury, only intoxicated patients presented with craniofacial and cervical spinal fractures during this period. Alcohol-related injuries are more often associated with head/neck fractures but less extremity injuries. The rate of fractures missed by emergency physicians but later diagnosed by radiologists was the same in intoxicated and sober patients.

Human papillomavirus-associated plantar epidermoid cysts: MR and US imaging appearance.

Epidermoid cysts are epithelial cysts that present as slow-growing intradermal or subcutaneous lesions. While recent epidemiological studies have isolated human papillomavirus (HPV) from plantar epidermoid cysts, imaging findings in HPV-associated epidermoid cysts have not been previously reported. We describe imaging findings in two patients with HPV-associated plantar epidermoid cysts. Magnetic resonance (MR) imaging and ultrasonography (US) showed linear arrangement of several adjacent globular cysts. This appearance is hypothesized to result from HPV-associated eccrine duct metaplasia leading to cyst formation and later traumatic rupture leading to formation of multiple adjacent cystic components. It may be useful to suggest assessing the presence of HPV antigen in plantar lesions having these imaging findings.

Imaging of vascular tumors with an emphasis on ISSVA classification.

The International Society for the Study of Vascular Anomalies (ISSVA) classification is becoming the international standard classification system for vascular tumors and vascular malformations. The ISSVA classification strictly distinguishes vascular tumors (neoplastic lesions) from vascular malformations (non-neoplastic lesions) based on whether there is a proliferation of vascular endothelial cells present, and it is an extremely useful classification system for determining therapeutic measures. For vascular tumors, it is clinically significant in terms of discriminating infantile hemangioma and rapidly involuting congenital hemangioma, which are expected to spontaneously regress, from other vascular tumors requiring treatment. Needless to say, clinical courses are important for diagnosis, and it is also important for radiologists to understand imaging findings on vascular tumors because such tumors have unique findings on diagnostic images. In this paper, vascular tumors are classified based on the ISSVA classification, and clinical and imaging findings are reviewed.

Magnetic resonance imaging for extramammary Paget's disease: radiological and pathological correlations.

Extramammary Paget's disease (EMPD) is a rare cutaneous neoplasm that is thought to represent intraepithelial adenocarcinoma developing in an area rich in apocrine glands. Magnetic resonance imaging (MRI) findings for this disease are not well established. We report three cases of pathologically confirmed EMPD in which MRI was performed before surgery. The lesions were widespread in the epidermis and the dermis. Lesions were sharply well enhanced on gadolinium-enhanced T1-weighted imaging and appeared hyperintense on diffusion-weighted imaging in all cases. Areas with enhancement in depth corresponded well with the pathological lesion. In addition, different malignant legions were found on the same images from MRI in two cases, indicating potential associations with other malignancies. We describe the MRI findings and their pathological correlation. MRI could be useful for preoperative evaluation of disease spread and detection of associated malignancies.

Severity of pseudofilling defect in the left atrial appendage on cardiac computed tomography is a simple predictor of the degree of left atrial emptying dysfunction in patients with chronic atrial fibrillation.

OBJECTIVES: The purpose of our study was to investigate the clinical relevance of a pseudofilling defect in the left atrial appendage (LAA) detected on coronary computed tomography (CT) angiography (CCTA) as an indicator of impaired left atrial (LA) volumetric function in patients with chronic atrial fibrillation (CAF). METHODS: Forty-two patients with CAF underwent CCTA. Quantitative and visual measurements of contrast enhancement of the LAA were performed, and they were correlated with results of CT volumetric functional analysis of the LA. Four volumetric parameters representing LA function were measured: maximum (LAVmax) and minimum volumes of the LA (LAVmin) through the entire cardiac cycle; LA emptying volume (LAEV); and LA emptying fraction (LAEF). All volumetric parameters were standardized by body surface area to adjust for variation in LA size among patients. For quantitative measurement, the CT attenuation was measured at the LAA and the LA to calculate an LAA/LA attenuation ratio. For visual measurement, contrast enhancement of the LAA was categorized into 3 groups; no filling defect, mild-to-moderate pseudofilling defect, and severe pseudofilling defect group. The Spearman correlation coefficient and the Kruskal-Wallis test were used for statistical analysis. RESULTS: The LAA/LA ratio showed a strong positive correlation with LAEV (r = 0.52; P < 0.001) and LAEF (r = 0.69; P < 0.001). The LAEV in the no pseudofilling defect group and the mild-to-moderate and severe pseudofilling defect groups were 16.1 ± 8.4, 10.8 ± 3.1, and 6.7 ± 4.9 mL/m², respectively (P < 0.001). The LAEF in each group were 24.2 ± 13.8%, 12.0 ± 3.4%, and 6.9 ± 3.1%, respectively (P < 0.001). CONCLUSIONS: The severity of pseudofilling defect in the LAA on CCTA could predict the degree of LA emptying dysfunction in patients with CAF.

Three-dimensional T2-weighted imaging using the dark blood method for detecting pulmonary embolisms: comparison with computed tomography angiography.

PURPOSE: The aim of this study was to assess the potential diagnostic value of three-dimensional T2-weighted imaging using the dark blood method for detecting pulmonary embolism (PE). MATERIALS AND METHODS: Ten consecutive patients with already diagnosed acute PE (five men, five women; mean age 58.6 years, range 35-79 years) were prospectively enrolled in this study. All patients underwent lung magnetic resonance imaging (MRI) with a 1.5-T scanner. MRI was performed on the same day as CT angiography (CTA), which was undertaken to monitor treatment. Two radiologists performed a consensus evaluation of MRI followed by CTA on a per-vessel basis: the main pulmonary artery (PA), lobar PA, segmental PA, and subsegmental PA. Each modality was evaluated independently on separate days (at least 8 weeks apart). The accuracy of MRI for detecting PE was determined by comparing it with CTA results, which were used as a reference standard. Cohen's kappa analysis was used for statistical analyses. RESULTS: Among the 10 patients, pulmonary emboli were seen in 6 of 20 main arteries, 22 of 60 lobar arteries, 35 of 180 segmental arteries, and 8 of 410 subsegmental arteries on CTA. The sensitivities/specificities of MRI were 100%/100% for the main PA, 90.9%/97.3% for the lobar PA, and 74.2%/97.9% for the segmental PA, respectively. Altogether, 304 (83%) of 410 subsegmental arteries were not visualized on MRI and only 1 of the 8 emboli in the subsegmental branches was depicted. The kappa values for the main, lobar, and segmental arteries were 1.0, 0.89, and 0.77, respectively. CONCLUSION: Three-dimensional T2-weighted imaging using the dark blood method appears to be principally useful for diagnosing main, lobar, and segmental PEs.