Search

Purpose: To distinguish malignant and benign bowel wall thickening (BWT) by using computed tomography (CT) texture features based on machine learning (ML) models and to compare its success with the clinical model and combined model., Methods: One hundred twenty-two patients with BWT identified on contrast-enhanced abdominal CT and underwent colonoscopy were included in this retrospective study. Texture features were extracted from CT images using LifeX software. Feature selection and reduction were performed using the Least Absolute Shrinkage and Selection Operator (LASSO). Six radiomic features were selected with LASSO. In the clinical model, six features (age, gender, thickness, fat stranding, symmetry, and lymph node) were included. Six radiomic and six clinical features were used in the combined model. Classification was done using two machine learning algorithms: Support Vector Machine (SVM) and Logistic Regression (LR). The data sets were divided into 80% training set and 20% test set. Then, training took place with all three datasets. The model's success was tested with the test set consisting of features not used during training., Results: In the training set, the combined model had the best performance with the area under the curve (AUC) value of 0.99 for SVM and 0.95 for LR. In the radiomic-derived model, the AUC value is 0.87 in SVM and 0.79 in LR. In the clinical model, SVM made this distinction with 0.95 AUC and LR with 0.92 AUC value. In the test set, the classifier with the highest success distinguishing malignant wall thickening is SVM in the radiomic-derived model with an AUC value of 0.90. In other models, the AUC value is in the range of 0.75-0.86, and the accuracy values are in the range of 0.72-0.84., Conclusion: In conclusion, radiomic-based machine learning has shown high success in distinguishing malignant and benign BWT and may improve diagnostic accuracy compared to clinical features only. The results of our study may help ensure early diagnosis and treatment of colorectal cancers by facilitating the recognition of malignant BWT., (© 2024. The Author(s) under exclusive licence to Japan Radiological Society.)

Objectives: The artificial intelligence competition in healthcare at TEKNOFEST-2022 provided a platform to address the complex multi-class classification challenge of abdominal emergencies using computer vision techniques. This manuscript aimed to comprehensively present the methodologies for data preparation, annotation procedures, and rigorous evaluation metrics. Moreover, it was conducted to introduce a meticulously curated abdominal emergencies data set to the researchers., Methods: The data set underwent a comprehensive central screening procedure employing diverse algorithms extracted from the e-Nabız (Pulse) and National Teleradiology System of the Republic of Türkiye, Ministry of Health. Full anonymization of the data set was conducted. Subsequently, the data set was annotated by a group of ten experienced radiologists. The evaluation process was executed by calculating F 1 scores, which were derived from the intersection over union values between the predicted bounding boxes and the corresponding ground truth (GT) bounding boxes. The establishment of baseline performance metrics involved computing the average of the highest five F 1 scores., Results: Observations indicated a progressive decline in F 1 scores as the threshold value increased. Furthermore, it could be deduced that class 6 (abdominal aortic aneurysm/dissection) was relatively straightforward to detect compared to other classes, with class 5 (acute diverticulitis) presenting the most formidable challenge. It is noteworthy, however, that if all achieved outcomes for all classes were considered with a threshold of 0.5, the data set's complexity and associated challenges became pronounced., Conclusion: This data set's significance lies in its pioneering provision of labels and GT-boxes for six classes, fostering opportunities for researchers., Clinical Relevance Statement: The prompt identification and timely intervention in cases of emergent medical conditions hold paramount significance. The handling of patients' care can be augmented, while the potential for errors is minimized, particularly amidst high caseload scenarios, through the application of AI., Key Points: • The data set used in artificial intelligence competition in healthcare (TEKNOFEST-2022) provides a 6-class data set of abdominal CT images consisting of a great variety of abdominal emergencies. • This data set is compiled from the National Teleradiology System data repository of emergency radiology departments of 459 hospitals. • Radiological data on abdominal emergencies is scarce in literature and this annotated competition data set can be a valuable resource for further studies and new AI models., (© 2023. The Author(s), under exclusive licence to European Society of Radiology.)

Purpose: To investigate whether texture analysis of primary colonic mass in preoperative abdominal computed tomography (CT) scans of patients diagnosed with colon cancer could predict tumor grade, T stage, and lymph node involvement using machine learning (ML) algorithms., Materials and Methods: This retrospective study included 73 patients diagnosed with colon cancer. Texture features were extracted from contrast-enhanced CT images using LifeX software. First, feature reduction was performed by two radiologists through reproducibility analysis. Using the analysis of variance method, the parameters that best predicted lymph node involvement, grade, and T stage were determined. The predictive performance of these parameters was assessed using Orange software with the k-nearest neighbor (kNN), random forest, gradient boosting, and neural network models, and their area under the curve values were calculated., Results: There was excellent reproducibility between the two radiologists in terms of 49 of the 58 texture parameters that were subsequently subject to further analysis. Considering all four ML algorithms, the mean AUC and accuracy ranges were 0.557-0.800 and 47-76%, respectively, for the prediction of lymph node involvement; 0.666-0.846 and 68-77%, respectively, for the prediction of grade; and 0.768-0.962 and 81-88%, respectively, for the prediction of T stage. The best performance was achieved with the random forest model in the prediction of LN involvement, the kNN model for the prediction of grade, and the gradient boosting model for the prediction of T stage., Conclusion: The results of this study suggest that the texture analysis of preoperative CT scans obtained for staging purposes in colon cancer can predict the presence of advanced-stage tumors, high tumor grade, and lymph node involvement with moderate specificity and sensitivity rates when evaluated using ML models., (© 2023. The Author(s) under exclusive licence to Japan Radiological Society.)

Objective: The artificial intelligence competition in healthcare was organized for the first time at the annual aviation, space, and technology festival (TEKNOFEST), Istanbul/Türkiye, in September 2021. In this article, the data set preparation and competition processes were explained in detail; the anonymized and annotated data set is also provided via official website for further research., Materials and Methods: Data set recorded over the period covering 2019 and 2020 were centrally screened from the e-Pulse and Teleradiology System of the Republic of Türkiye, Ministry of Health using various codes and filtering criteria. The data set was anonymized. The data set was prepared, pooled, curated, and annotated by 7 radiologists. The training data set was shared with the teams via a dedicated file transfer protocol server, which could be accessed using private usernames and passwords given to the teams under a nondisclosure agreement signed by the representative of each team., Results: The competition consisted of 2 stages. In the first stage, teams were given 192 digital imaging and communications in medicine images that belong to 1 of 3 possible categories namely, hemorrhage, ischemic, or non-stroke. Teams were asked to classify each image as either stroke present or absent. In the second stage of the competition, qualifying 36 teams were given 97 digital imaging and communications in medicine images that contained hemorrhage, ischemia, or both lesions. Among the employed methods, Unet and DeepLabv3 were the most frequently observed ones., Conclusion: Artificial intelligence competitions in healthcare offer good opportunities to collect data reflecting various cases and problems. Especially, annotated data set by domain experts is more valuable.

Purpose: Benign essential blepharospasm (EB) is a focal facial dyskinesia that occurs with the involuntary contraction of muscles around the eyes. In the literature, studies on blepharospasm focus on elucidating the pathophysiology of this condition in the brain. To the best of our knowledge, there is no research evaluating the orbital imaging findings of patients with EB. Therefore, the current study aimed to determine whether there was any change in the blood supply of muscles around the eye or ocular blood flow in patients with blepharospasm and investigate flow changes that may be caused by spasm., Methods: Thirty patients with EB and 30 controls were included in the study. Orbital Doppler ultrasound was performed in all cases to measure ophthalmic and temporal artery peak systolic velocity and end diastolic velocity and calculate resistive index values. Superior ophthalmic vein blood flow velocity (SOVBFV) was also measured., Results: There was no significant difference between the groups in terms of age and gender distribution (P = 0.345 and 0.870, respectively). SOVBFV was found to be significantly higher in the EB group (P = 0.001). No significant difference was observed in the remaining investigated parameters., Conclusions: In conclusion, our study suggested that ocular spasm in patients with EB had no effect on blood flow other than SOVBFV. When SOVBFV was compared between the EB and control groups, it was found to be increased in the EB group. We consider that this statistical difference may be clinically and pathophysiologically significant., (© 2021. Società Italiana di Ultrasonologia in Medicina e Biologia (SIUMB).)

Background: With advances in surgical techniques and immunosuppression, liver transplantation has become the most effective treatment of acute and chronic liver failures. Evaluation of vascular anatomy and detection of hepatic vascular variations prior to surgery, especially transplantation surgery, can help reduce complications in both the donor and the recipient. Intraoperative ultrasonography (IOUS) is known to be beneficial during planning of the transplantation surgery, and can help direct the surgery itself., Objectives: To our knowledge, there are no existing studies that evaluate the number and diameter of segment 5 and 8 branches that need to be anastomosed with IOUS., Patients and Methods: In this study, considering surgical anatomical evaluation as the gold standard, IOUS findings were compared to computed tomography angiography (CTA) findings. 40 patients were included in the study., Results: The average diameters of segment 8 branches that were anastomosed and not anastomosed were significantly different when measured by IOUS (p = 0.016); however, no such statistically significant difference was found in measurements made with CTA (p = 0.89)., Conclusion: CTA is superior to IOUS in detecting segment 5 and 8 veins draining into the middle hepatic vein. However, IOUS is more accurate in predicting which vessels are going to be anastomosed. For a complete and accurate assessment, both imaging modalities should be used to complement each other, and their respective advantages and disadvantages should be known., (© 2021. Società Italiana di Ultrasonologia in Medicina e Biologia (SIUMB).)

Purpose: The aim of this study was to investigate the efficacy of shear wave elastography (SWE) in the diagnosis of perforating vein insufficiency, and to determine the applicability of these measurements., Methods: A total of 140 symptomatic patients with a total of 280 lower extremities were investigated. All patients presented with venous insufficiency (VI) symptoms, and received Doppler ultrasound assessment to determine VI and SWE measurements. The SWE values were measured in the adjacent perivenous tissue of the largest Cockett's perforating vein (PV) of both lower extremities, at the level where they pass the fascia. The Cockett's PV diameter and the presence of reflux in Cockett's PV and the great saphenous vein were compared with SWE values in perivenous tissue of PVs., Results: The SWE values of the perforating vein insufficiency group were significantly higher than those of the normal PV without insufficiency group (P < 0.001). A significant and positive relation was seen between increased PV diameter and SWE values (P < 0.001) and there was a significant relationship between the presence of perforating vein insufficiency and increase in PV diameter. A statistically significant increase was detected in SWE values for the PV for those with reflux in the great saphenous vein (P < 0.001). The best cut-off values that can be used to detect perforating vein insufficiency were found 34.600 for kPa and 3.375 for m/s., Conclusion: SWE may be used effectively in addition to conventional Doppler ultrasound examination in diagnosing and following perforating vein insufficiency., (© 2020. Società Italiana di Ultrasonologia in Medicina e Biologia (SIUMB).)

One of the rarely reported computed tomography findings of COVID-19 is an air bubble sign. Minimum intensity projection (MinIP) images can increase the prominence of the air bubble appearance, which is one of the low-density findings observed in COVID-19 pneumonia. We present two cases with COVID-19 pneumonia, who were not optimally selected on thoracic tomography images, and air bubble appearance was detected in MinIP images. While air bubbles are not noticed in normal thin-section computed tomography images, they become more easily noticeable after the MinIP technique is applied. This observation needs to be confirmed by radiological studies.

One of the major problems radiologists face in everyday practice is to decide the correct diagnosis, or at least narrow down the list of possibilities. In this context, indicative evidences (signs) are useful to recognize pathologies, and also to narrow the list of differential diagnoses. Despite classically being described for a single disease, or a closely related family of disorders, most indications are not restricted exclusively to their traditional definition. Therefore, using signs for prognosis requires knowledge of the mechanism of their appearance, and which pathologies they are observed in. In this study, we demonstrate some of the more common and useful neuroradiologic signs with relevant images, and discuss their use in differential diagnosis.

Purpose: We aimed to investigate the spectrum of radiologic findings and referral reasons for breast diseases in children considering age-appropriate presentation., Methods: Our retrospective cohort study included 348 consecutive pediatric patients aged <19 years (median, 13 years) referred to radiology with a clinical presentation between 2005 and 2016. Radiologic findings were reviewed in four age ranges (0-2 years, 2-8 years, 8-15 years, >15 years)., Results: Of 348 patients, 257 had a referral reason. The most frequent referral reason was a palpable mass (35%). Developmental abnormalities accounted for 48% of all radiologic findings in 348 patients. We did not detect any breast malignancy. According to age groups, the most common radiologic findings were neonatal hypertrophy (0-2 years), early breast development (2-8 years), developmental abnormalities by a majority of gynecomastia (8-15 years), and normal findings or developmental abnormalities (>15 years). Interestingly, the frequency of gynecomastia was only 4% in neonatal period or early childhood. Fibroadenomas and fibroadenoma-like solid masses were seen after 8 years and constituted the majority of solid masses (65%). Cysts were seen at a rate of 7% and majority of them were of simple type, which tends to resolve in time., Conclusion: In our study, the most common referral reason to radiology was a palpable breast mass. Neonatal hypertrophy and early breast development in younger children, and developmental abnormalities in older children may be kept in mind as the most common radiologic findings. Our study confirms the substantial absence of malignancies in children as well as a widely different disease spectrum in comparison with the adult population.

Aims: Intrathyroidal ectopic thymus (IET) is being increasingly reported in the radiology literature. Most of the reports are of individual cases or small series and prevalence and natural course of the pathology is not well known. The purpose of this study is to establish the prevalence of IET in children and report long term follow-up results., Material and Methods: In 180 children who were examined by ultrasound (US) for other reasons, 7 patients were indentified with IET. Together with the other seven children who were already under follow-up for IET (diagnosed using US criteria), these 14 patients were followed up with US for 30 months. Size, shape, location, echotexture and internal echoes of the lesions were evaluated., Results: There were 16 lesions in 14 children. The most common appearance was a fusiform hypoechoic lesion, with punctate and linear internal echoes and well-defined but slightly irregular borders located posteriorly in the lower thirds of the thyroid. In follow-up, there were no changes in echotexture, shape or border. In 3 patients, the lesion became slightly smaller, in a 10-year-old boy slightly larger, and in an 11-year old boy the lesion disappeared. In a patient with bilateral lesions, one lesion slightly decreased in size., Conclusions: IET in children may be more common than thought. Its growth reflects that of a normal thymus. Awareness of this entity is important in order not to misdiagnose them, especially as papillary cancer, and to prevent unnecessary interventions.