Insights and Considerations in Development and Performance Evaluation of Generative Adversarial Networks (GANs): What Radiologists Need to Know.

The rapid development of deep learning in medical imaging has significantly enhanced the capabilities of artificial intelligence while simultaneously introducing challenges, including the need for vast amounts of training data and the labor-intensive tasks of labeling and segmentation. Generative adversarial networks (GANs) have emerged as a solution, offering synthetic image generation for data augmentation and streamlining medical image processing tasks through models such as cGAN, CycleGAN, and StyleGAN. These innovations not only improve the efficiency of image augmentation, reconstruction, and segmentation, but also pave the way for unsupervised anomaly detection, markedly reducing the reliance on labeled datasets. Our investigation into GANs in medical imaging addresses their varied architectures, the considerations for selecting appropriate GAN models, and the nuances of model training and performance evaluation. This paper aims to provide radiologists who are new to GAN technology with a thorough understanding, guiding them through the practical application and evaluation of GANs in brain imaging with two illustrative examples using CycleGAN and pixel2style2pixel (pSp)-combined StyleGAN. It offers a comprehensive exploration of the transformative potential of GANs in medical imaging research. Ultimately, this paper strives to equip radiologists with the knowledge to effectively utilize GANs, encouraging further research and application within the field.

Rapidly growing distal choroidal artery aneurysm re-rupture following revascularization for hemorrhagic Moyamoya disease: A case report.

Intracranial hemorrhage is the leading cause of neurological deficits and poor prognosis in adult patients with Moyamoya disease (MMD). Intracranial hemorrhage is occasionally accompanied by MMD-associated aneurysm and requires additional treatment. To date, direct or indirect bypass surgery or endovascular treatment, such as coil embolization, has been adopted and has achieved successful outcomes. The rapid growth of MMD-associated aneurysms and rebleeding after direct bypass surgery via superficial temporal artery-middle cerebral artery (STA-MCA) anastomosis has rarely been reported. We report a case of a rapidly growing fragile arterial pseudoaneurysm in a patient with MMD. A 45-year-old female was admitted with a headache and decreased mental status. Radiological evaluation, including distal subtraction angiography, revealed intraventricular hemorrhage with a left posterior choroidal artery pseudoaneurysm. Within 4 days after revascularization surgery via STA-MCA direct bypass, the size of the pseudoaneurysm rapidly increased and rebleeding occurred, requiring coil embolization. After endovascular therapy and a second STA-MCA bypass surgery, the patient recovered well and was discharged 8 days later. Follow-up radiological imaging revealed an obliterated pseudoaneurysm without rebleeding or complications. In this case, the rapid growth of an MMD-associated pseudoaneurysm was observed after revascularization surgery because of temporary hemodynamic instability. This report raises questions regarding the causes and management of unstable postbypass hemodynamics.

Dural tail sign positive tumors: Points to make a differential diagnosis.

Representative patients were treated with total surgical mass resection, and each tumor was histopathologically confirmed to have a secretory meningioma, intradural metastasis of gynecologic origin, and dural metastasis of lung origin. The imaging findings of these patients were inconclusive in differentiating meningioma from metastasis; hence, advanced magnetic resonance imaging (MRI) techniques were considered. Based on these reports, we studied how to differentiate typical meningiomas from atypical and malignant meningiomas and other dura-based malignant tumors using conventional computed tomography and MRI.

Deep learning-based detection algorithm for brain metastases on black blood imaging.

Brain metastases (BM) are the most common intracranial tumors, and their prevalence is increasing. High-resolution black-blood (BB) imaging was used to complement the conventional contrast-enhanced 3D gradient-echo imaging to detect BM. In this study, we propose an efficient deep learning algorithm (DLA) for BM detection in BB imaging with contrast enhancement scans, and assess the efficacy of an automatic detection algorithm for BM. A total of 113 BM participants with 585 metastases were included in the training cohort for five-fold cross-validation. The You Only Look Once (YOLO) V2 network was trained with 3D BB sampling perfection with application-optimized contrasts using different flip angle evolution (SPACE) images to investigate the BM detection. For the observer performance, two board-certified radiologists and two second-year radiology residents detected the BM and recorded the reading time. For the training cohort, the overall performance of the five-fold cross-validation was 87.95%, 24.82%, 19.35%, 14.48, and 18.40 for sensitivity, precision, F1-Score, the false positive average for the BM dataset, and the false positive average for the normal individual dataset, respectively. For the comparison of reading time with and without DLA, the average reading time was reduced by 20.86% in the range of 15.22-25.77%. The proposed method has the potential to detect BM with a high sensitivity and has a limited number of false positives using BB imaging.