Search

The purpose of this study was to reveal the usefulness of machine learning classifier and feature selection algorithms for prediction of insufficient hepatic enhancement in the HBP. We retrospectively assessed 214 patients with chronic liver disease or liver cirrhosis who underwent MRI enhanced with Gd-EOB-DTPA. Various liver function tests, Child–Pugh score (CPS) and Model for End-stage Liver Disease Sodium (MELD-Na) score were collected as candidate predictors for insufficient hepatic enhancement. Insufficient hepatic enhancement was assessed using liver-to-portal vein signal intensity ratio and 5-level visual grading. The clinico-laboratory findings were compared using Student’s t-test and Mann–Whitney U test. Relationships between the laboratory tests and insufficient hepatic enhancement were assessed using Pearson’s and Spearman’s rank correlation coefficient. Feature importance was assessed by Random UnderSampling boosting algorithms. The predictive models were constructed using decision tree(DT), k-nearest neighbor(KNN), random forest(RF), and support-vector machine(SVM) classifier algorithms. The performances of the prediction models were analyzed by calculating the area under the receiver operating characteristic curve(AUC). Among four machine learning classifier algorithms using various feature combinations, SVM using total bilirubin(TB) and albumin(Alb) showed excellent predictive ability for insufficient hepatic enhancement(AUC = 0.93, [95% CI: 0.93–0.94]) and higher AUC value than conventional logistic regression(LR) model (AUC = 0.92, [95% CI; 0.92–0.93], predictive models using the MELD-Na (AUC = 0.90 [95% CI: 0.89–0.91]) and CPS (AUC = 0.89 [95% CI: 0.88–0.90]). Machine learning-based classifier (i.e. SVM) and feature selection algorithms can be used to predict insufficient hepatic enhancement in the HBP before performing MRI.

To evaluate the association between thyroid echogenicity and heterogeneity seen on ultrasonography (US) and thyroid function in pediatric and adolescent populations with autoimmune diffuse thyroid diseases (AITD). From 2000 to 2020, we reviewed thyroid ultrasound (US) images and thyroid function statuses in 133 children and adolescent AITD patients. Our review of the images focused on decreased echogenicity and heterogeneity, which were classified into four grades. Among patients with overt hypothyroidism or overt hyperthyroidism, 94.2% (65/69) showed a US grade of 3 or 4. In patients with subclinical hyper/hypothyroidism or euthyroidism, 45.3% (29/64) showed grades 1 or 2. There were no overt hyper/hypothyroidism patients with US grade 1. When we compared US grades according to thyroid status, more severe thyroid dysfunction was significantly associated with higher US grade (p = 0.047). Thyroid stimulating hormone (TSH) level differed significantly according to US grades when we evaluated hyperthyroid (p = 0.035) and hypothyroid (p = 0.027) states independently. 11 patients showed both US grade and thyroid function status changes on follow-up US. In children and adolescent AITD patients, there was an association between decreased echogenicity and heterogeneity on US and thyroid dysfunction.

Purpose We investigated whether [18F]fluorodeoxyglucose ([18F]FDG) positron emission tomography/computed tomography (PET/CT) taken in the prone position could reduce [18F]FDG uptake in dependent lungs. Methods Patients who underwent [18F]FDG PET/CT in both supine and prone positions from October 2018 to September 2021 were reviewed retrospectively. [18F]FDG uptake of dependent and nondependent lungs was analysed visually and semi-quantitatively. A linear regression analysis was also performed to examine the association between the mean standardised uptake value (SUVmean) and the Hounsfield unit (HU). Results Totally, 135 patients (median age, 66 years [interquartile range: 58–75 years]; 80 men) were included. Dependent lungs showed significantly higher SUVmean and HU than nondependent lungs on both supine position PET/CT (sPET/CT, 0.59 ± 0.14 vs. 0.36 ± 0.09, p p p p mean and HU in sPET/CT (R = 0.86, p p 18F]FDG uptake in the posterior lung on sPET/CT, which disappeared on pPET/CT in all but one patient (0.7%, p 18F]FDG uptake of the lung had moderate-to-strong associations with HU. Gravity-dependent opacity-related [18F]FDG uptake can be effectively reduced on prone position PET/CT.

2-Deoxy-2-[18F]fluoro-D-glucose ([18F]FDG) positron-emission tomography/computed tomography (PET/CT) is widely used to evaluate lung nodules, although respiratory motion artefacts may occur. We investigated the value of prone position PET/CT (pPET/CT) in lung nodule evaluation compared with standard supine position PET/CT (sPET/CT). We retrospectively reviewed 28 consecutive patients (20 men; age, 65.6 ± 12.1 years) with a lung nodule (size, 16.8 ± 5.5 mm) located below the sub-carinal level who underwent [18F]FDG PET/CT in a standard supine position and additional prone position. The maximum standardised uptake value (SUVmax), metabolic tumour volume (MTV), difference of diaphragm position between PET and CT (DDP), Dice’s similarity coefficient (DSC) and occurrence of mis-registration were analysed. The [18F]FDG uptake of 20 biopsy-confirmed (15 malignant) nodules was evaluated visually. pPET/CT yielded a significantly higher SUVmax, lower MTV and shorter DDP than with sPET/CT (p = 0.043, 0.007 and 0.021, respectively). Mis-registration occurred in 53.6% of cases in sPET/CT and in 28.6% of cases in pPET/CT (p = 0.092). Among the 15 patients with mis-registration in sPET/CT, 10 patients (66.7%) did not show mis-registration in pPET/CT. DSC was higher in pPET/CT than in sPET/CT in 18 out of 28 patients (64.3%). In visual analysis, malignant nodules exhibited a higher [18F]FDG uptake positivity than benign nodules in pPET/CT (93.3% vs. 40.0%, p = 0.032) but not in sPET/CT (80.0% vs. 40.0%, p = 0.131). pPET/CT reduces respiratory motion artefact and enables more-precise measurements of PET parameters. • In prone position PET/CT, the decrease in the blurring effect caused by reduced respiratory motion resulted in a higher SUV max and lower MTV in lung nodules than that with supine position PET/CT. • Prone position PET/CT was useful to interpret correctly malignant lung nodules as being positive in individual cases that had a negative result in supine position PET/CT.

Objectives: The prone position is useful in reducing respiratory motion artifacts in lung nodules on 2-Deoxy-2-[18F]fluoro-D-glucose ([18F]FDG) positron-emission tomography/computed tomography (PET/CT). However, whether prone position PET/CT (pPET/CT) is useful in evaluating hepatic lesions is unknown. Methods: Thirty-five hepatic lesions (size, 13.0 ± 5.8 mm) from 20 consecutive patients (age, 67.0 ± 9.7 years) were evaluated. The maximum standardized uptake value (SUVmax) and metabolic tumor volume (MTV) of both pPET/CT and standard supine position PET/CT (sPET/CT) were evaluated. Results: No significant difference in SUVmax (4.41 ± 2.0 vs. 4.23 ± 1.83; p = 0.240) and MTV (5.83 ± 6.69 vs. 5.95 ± 6.24; p = 0.672) was observed between sPET/CT and pPET/CT. However, SUVmax changes on pPET/CT varied compared with those on sPET/CT (median, −4%; range: −30%–71%). pPET/CT was helpful when the [18F]FDG uptake of the hepatic lesions was located outside the liver on sPET/CT (n = 4, SUVmax change: median 15%; range: 7%–71%) and more severe blurring on sPET/CT (n = 6, SUVmax change: median 11%; range: −3%–32%). Conclusions:The SUVmax of hepatic lesions varies on pPET/CT compared with that on sPET/CT, and pPET/CT may be helpful in individual cases.

This study was conducted to investigate and compare thyroid ultrasonography (US) findings in children and adolescents with goiter and normal thyroid function with positive or negative thyroid autoimmunity. From 2000 to 2020, we reviewed initial thyroid US images in 33 autoimmune thyroid diseases (AITDs) patients and 52 nonAITD patients. Our review of the images focused on thyroid parenchymal hypoechogenicity and heterogeneous echopattern subdivided into 2 groups according to severity: hypoechogenicity 1 and 2 (HO1 and HO2) and heterogeneity 1 and 2 (HE1 and HE2). HO1 and HE1 were observed more frequently in the nonAITD group (86.5% and 42.3%, respectively), while HO2 and HE2 were observed more frequently in the AITDs group (36.4% and 81.8%, respectively). More patients in the AITDs group showed change of both US groups and thyroid function state within the follow-up periods than in nonAITD group (33.3% and 5.77%, respectively). Children and adolescent AITDs patients showed more severe parenchyma hypoechogenicity and heterogeneous echopattern compared with nonAITD patients with goiter and normal thyroid function.

Necrotizing fasciitis (NF) is a life-threatening infection characterized by extensive necrosis and inflammation of subcutaneous tissue and the fascia. Only a few cases of NF in the breast have been reported, and imaging findings of primary breast NF have not been described in the literature. As primary NF in the breast is extremely rare, it can be misdiagnosed as an abscess or cellulitis, and its diagnosis may be delayed. However, early diagnosis is crucial because delays can lead to fatal sepsis or requirement for total mastectomy. Herein, the authors report a rare case of primary breast NF that was diagnosed early using enhanced breast CT and successfully managed with local debridement. CT revealed a large cystic mass with an air-fluid level, a thickened deep fascia without remarkable enhancement, and extensive subcutaneous emphysema with subcutaneous fat infiltrations in the right breast.

The purpose of this study was to reveal the usefulness of machine learning classifier and feature selection algorithms for prediction of insufficient hepatic enhancement in the HBP.We retrospectively assessed 214 patients with chronic liver disease or liver cirrhosis who underwent MRI enhanced with Gd-EOB-DTPA. Various liver function tests, Child-Pugh score (CPS) and Model for End-stage Liver Disease Sodium (MELD-Na) score were collected as candidate predictors for insufficient hepatic enhancement. Insufficient hepatic enhancement was assessed using liver-to-portal vein signal intensity ratio and 5-level visual grading. The clinico-laboratory findings were compared using Student's t-test and Mann-Whitney U test. Relationships between the laboratory tests and insufficient hepatic enhancement were assessed using Pearson's and Spearman's rank correlation coefficient. Feature importance was assessed by Random UnderSampling boosting algorithms. The predictive models were constructed using decision tree(DT), k-nearest neighbor(KNN), random forest(RF), and support-vector machine(SVM) classifier algorithms. The performances of the prediction models were analyzed by calculating the area under the receiver operating characteristic curve(AUC).Among four machine learning classifier algorithms using various feature combinations, SVM using total bilirubin(TB) and albumin(Alb) showed excellent predictive ability for insufficient hepatic enhancement(AUC = 0.93, [95% CI: 0.93-0.94]) and higher AUC value than conventional logistic regression(LR) model (AUC = 0.92, [95% CI; 0.92-0.93], predictive models using the MELD-Na (AUC = 0.90 [95% CI: 0.89-0.91]) and CPS (AUC = 0.89 [95% CI: 0.88-0.90]).Machine learning-based classifier (i.e. SVM) and feature selection algorithms can be used to predict insufficient hepatic enhancement in the HBP before performing MRI.

Purpose: To evaluate the association between thyroid echogenicity and heterogeneity seen on ultrasonography (US) and thyroid function in pediatric and adolescent populations with autoimmune diffuse thyroid diseases (AITD).Methods: From 2000 to 2020, we reviewed thyroid ultrasound (US) images and thyroid function statuses in 133 children and adolescent AITD patients. Our review of the images focused on decreased echogenicity and heterogeneity, which were classified into four grades.Results: Among patients with overt hypothyroidism or overt hyperthyroidism, 94.2% (65/69) showed a US grade of 3 or 4. In patients with subclinical hyper/hypothyroidism or euthyroidism, 45.3% (29/64) showed grades 1 or 2. There were no overt hyper/hypothyroidism patients with US grade 1. When we compared US grades according to thyroid status, more severe thyroid dysfunction was significantly associated with higher US grade (p=0.047). Thyroid stimulating hormone (TSH) level differed significantly according to US grades when we evaluated hyperthyroid (p=0.035) and hypothyroid (p=0.027) states independently. 11 patients showed both US grade and thyroid function status changes on follow-up US.Conclusions: In children and adolescent AITD patients, there was an association between decreased echogenicity and heterogeneity on US and thyroid dysfunction.

We retrospectively assessed 214 patients with chronic liver disease or liver cirrhosis who underwent magnetic resonance imaging (MRI) enhanced with gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid (Gd-EOB-DTPA) from August 2016 to May 2020 to evaluate the relationship between biochemical results that reflect liver function and hepatic enhancement. With the information gained we employed a machine learning approach with the K-Nearest Neighbor (KNN) algorithm to develop a predictive model for determining insufficient hepatic enhancement during the hepatobiliary phase (HBP) in Gd-EOB-DTPA-enhanced MRI. Using both quantitative and qualitative assessments, the total bilirubin (TB), albumin (Alb), prothrombin time-international normalized ratio, platelet, Child-Pugh score (CPS), and Model for End-stage Liver Disease Sodium (MELD-Na) score were related to decreased hepatic enhancement. In a multivariate analysis, TB and Alb were associated with insufficient enhancement (p

2-Deoxy-2-[We retrospectively reviewed 28 consecutive patients (20 men; age, 65.6 ± 12.1 years) with a lung nodule (size, 16.8 ± 5.5 mm) located below the sub-carinal level who underwent [pPET/CT yielded a significantly higher SUVpPET/CT reduces respiratory motion artefact and enables more-precise measurements of PET parameters.• In prone position PET/CT, the decrease in the blurring effect caused by reduced respiratory motion resulted in a higher SUV

Background The maximum value of the strain ratio (SR) is a newly developed measure in strain-elastography. Purpose To prospectively compare the diagnostic performance of three different measures of strain-elastography, the maximum value of the SR (SRmax), the average value of the SR (SRave), and the color map, for differentiating benign and malignant breast lesions. Material and Methods We obtained the SRmax and SRave of 314 lesions from 290 patients with the tissue to nodule SR and color map using a five-degree scoring system. The diagnostic performances of the SRmax, SRave, and color map were compared after obtaining the area under the receiver operating characteristic (ROC) curves (AUCs) of each parameter. Results The AUC of the SRmax (0.7674) was larger than the AUCs of the SRave (0.7138) and color map (0.6324), with statistical significance ( P = 0.0383 for SRmax vs. SRave, P = 0.0000 for SRmax vs. color map). The AUC of the SRave was larger than that of the color map; however, there was no significant difference. The optimal cut-off point of the SRmax that balanced the sensitivity (91.12%) and specificity (50.81%) was 5.16. Conclusion The SRmax is a more reliable diagnostic tool than the SRave and color map for differentiating benign and malignant breast lesions.

This study aimed to investigate whether a deep learning reconstruction (DLR) method improves the image quality, stent evaluation, and visibility of the valve apparatus in coronary computed tomography angiography (CCTA) when compared with filtered back projection (FBP) and hybrid iterative reconstruction (IR) methods.CCTA images of 51 patients (mean age ± standard deviation [SD], 63.9 ± 9.8 years, 36 male) who underwent examination at a single institution were reconstructed using DLR, FBP, and hybrid IR methods and reviewed. CT attenuation, image noise, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and stent evaluation, including 10%-90% edge rise slope (ERS) and 10%-90% edge rise distance (ERD), were measured. Quantitative data are summarized as the mean ± SD. The subjective visual scores (1 for worst -5 for best) of the images were obtained for the following: overall image quality, image noise, and appearance of stent, vessel, and aortic and tricuspid valve apparatus (annulus, leaflets, papillary muscles, and chordae tendineae). These parameters were compared between the DLR, FBP, and hybrid IR methods.DLR provided higher Hounsfield unit (HU) values in the aorta and similar attenuation in the fat and muscle compared with FBP and hybrid IR. The image noise in HU was significantly lower in DLR (12.6 ± 2.2) than in hybrid IR (24.2 ± 3.0) and FBP (54.2 ± 9.5) (DLR reconstruction provided better images than FBP and hybrid IR reconstruction.

Summary Background CT radiation is arguably carcinogenic. Results from single-centre studies, mostly retrospective, have advocated lowering the CT radiation dose for the diagnosis of appendicitis. However, adoption of low-dose CT has been slow. We aimed to assess the effectiveness of low-dose CT compared with standard-dose CT in the diagnosis of appendicitis in adolescents and young adults. Methods We did this pragmatic, multicentre, randomi s ed controlled non-inferiority trial at 20 South Korean teaching hospitals with little experience with low-dose CT. Patients aged 15–44 years with suspected appendicitis were randomly assigned (1:1), via computer-generated random assignments (permuted block sizes of two, four, six, and eight) concealed in sequentially numbered envelopes, to receive low-dose CT (2 mSv) or standard-dose CT (≤8 mSv). Randomisation was stratified by site. Group allocation was concealed from patients, outcome assessors, and adverse event adjudicators; care providers, site pathologists, and data collectors were aware of allocation. The primary endpoint was the negative (unnecessary) appendectomy rate among all appendectomies, with a non-interiority margin of 4·5% for low-dose versus standard-dose CT. Primary analy s is was by modified intention to treat, which included all patients who received an appendectomy in the group to which they were assigned. This trial is registered with ClinicalTrials.gov, number NCT01925014. Findings Between Dec 4, 2013, and Aug 18, 2016, we assigned 1535 patients to the low-dose CT group and 1539 patients to the standard-dose CT group. 22 (3·9%) of 559 patients had a negative appendectomy in the low-dose group versus 16 (2·7%) of 601 patients in the standard-dose group (difference 1·3%, 95% CI −0·8 to 3·3; p=0·0022 for the non-inferiority test). We recorded 43 adverse events in 43 (2·8%) of 1535 patients in the low-dose group and 41 adverse events in 40 (2·6%) of 1539 patients in the standard-dose group. One life-threatening adverse event of anaphylaxis caused by an iodinated contrast material occurred in the low-dose group. Interpretation Radiation dose of appendiceal CT for adolescents and young adults can be reduced to 2 mSv without impairing clinical outcomes. In view of the vast number of appendiceal CT examinations done worldwide, use of low-dose CT could prevent a sizeable number of radiation-associated cancers in the future. Funding Korea Health Industry Development Institute, Seoul National University Bundang Hospital, Dasol Life Science, and Bracco Imaging Korea.

To assess the value of gadoxetic acid-enhanced and diffusion-weighted (DW) magnetic resonance (MR) imaging for differentiating primary hepatic angiosarcomas from hemangiomatosis and epithelioid hemangioendotheliomas (EHEs). We reviewed MR images of seven patients with pathologically determined hepatic angiosarcomas, 11 patients with hemangiomatosis, and five patients with EHEs. Two radiologists assessed morphologic features, signal intensity (SI), enhancement patterns, and the presence of diffusion restriction by consensus and compared between angiosarcoma vs hemangiomatosis and angiosarcoma vs EHEs. Angiosarcomas more frequently showed mixed well- and ill-defined margins (6, 85.7%), mixed strong and intermediate-high SI (5, 71.4%) on T2-weighted images, mixed peripheral and/or central nodular and rim and/or target enhancement (5, 71.4%), and mixed presence and absence of diffusion restriction (7, 100%) compared with hemangiomatosis and EHEs (P

AIM To investigate the characteristic radiologic findings of cryptogenic multifocal ulcerous stenosing enteritis (CMUSE) which can be differentiated from other similar bowel disease and to assess their clinical behavior. METHODS Twenty pathologically and clinically confirmed CMUSE patients (males:females = 8:12; mean age: 40.4 years) between March 2002 and August 2015 from seven academic centers in South Korea were retrospectively reviewed. We evaluated small bowel series (SBS; n = 25), computed tomography (CT) enterography (n = 21), magnetic resonance (MR) enterography (n = 2), and abdominopelvic CT (n = 18) images, focusing on enteric and perienteric manifestations. Any change in radiologic features during follow-up period was recorded. We evaluated clinical data including presenting symptoms, laboratory finding and presence of relapse from electronic medical records. Histopathologic findings were also evaluated. RESULTS The main symptoms were abdominal pain (n = 12) and anemia (n = 10). All patients showed small bowel strictures (n = 52, mean: 2.6 per patient) on initial CT/MR, located in the ileum (n = 47) or jejunum (n = 5). Strictures showed short-length (mean: 10.44 mm) and circumferential bowel wall thickening (mean: 5.56 mm) with layered enhancement (n = 48) that were also noted on initial SBS (n = 36) with shallow ulcers (n = 10). Some ulcerative lesions or wall thickening progressed into strictures on follow-up SBS/CT, and some strictures revealed recurrent ulceration on follow-up SBS. There were no penetrating disease features like fistula or abscess and no gastrointestinal tract involvement except the small bowel. Nine patients experienced disease recurrence (median relapse-free period: 32 mo) even post-operatively. Histopathologic features of surgically resected specimens were characterized as multiple superficial ulcerations confined to mucosa or submucosa and multiple strictures. CONCLUSION Under characteristic radiologic findings with multiple short-segmental strictures and/or shallow ulcers of the small intestine, CMUSE should be considered when assessing patients with recurrent abdominal pain and anemia.

Developmental venous anomaly (DVA) is a common congenital venous malformation characterized by dilated medullary veins in caput medusa configuration and a draining vein. Despite the high incidence of DVAs, they are benign anatomic variations and rarely cause symptoms. Here, we report computed tomography and magnetic resonance imaging findings with perfusion images of acute infarction from underlying DVA in a 63-year-old female patient who presented with acute onset of neurologic symptoms and recovered without any neurologic deficit.

High-resolution rectal MRI plays a crucial role in evaluating rectal cancer by providing multiple prognostic findings and imaging features that guide proper patient management. Quality reporting is critical for accurate effective communication of the information among multiple disciplines, for which a systematic structured approach is beneficial. Existing guides on reporting of rectal MRI are divergent on some issues, largely reflecting the differences in overall management of rectal cancer patients between the United States and Europe. The Korean Society of Abdominal Radiology (KSAR) study group for rectal cancer has developed an expert consensus recommendation regarding essential items for structured reporting of rectal cancer MRI using a modified Delphi method. This recommendation aims at presenting an up-to-date, evidence-based, practical, structured reporting template that can be readily adopted in daily clinical practice. In addition, a thorough explanation of the clinical and scientific rationale underlying the reporting items and their formats is provided. This KSAR recommendation may serve as a useful tool to help achieve more standardized optimal care for rectal cancer patients using rectal MRI.

Epithelioid angiomyolipoma (EAML) is an uncommon variant type of angiomyolipoma (AML), occurring primarily in the kidney and characterized by a distinct entity with a possibility of metastatic transformation and distant spread (1, 2). Although the imaging findings of classical AML are well described in the literature, there are only few case reports describing the imaging appearances of EAML especially originating from the retroperitoneum (1-3). Herein, we reported a case of perirenal EAML with computed tomography (CT) and ultrasonography (US) findings. CASE REPORT

Video-assisted thoracic surgery (VATS) is a minimally invasive technique commonly used worldwide for biopsy or resection of pulmonary lesions. Although VATS procedures are considered safe and effective, they have potential complications such as prolonged air leak, bleeding, infection, and port site metastasis. Port site metastasis is a rare complication after VATS for a malignancy, with a few reports of mesothelioma cases and others (1-4). However, to our knowledge, it has not been reported in a breast cancer patient in the English literature. We reported the first case of iatrogenic port site metastasis in a breast cancer patient who received VATS wedge resection for pulmonary metastasis. CASE REPORT